national height modernization studynational height modernization study national geodetic surveyi...

National Height Modernization Study

Report to Congress

U.S. DEPARTMENT OF COMMERCENational Oceanic and Atmospheric Administration

National Ocean ServiceNational Geodetic Survey

June 1998

William M. DaleySecretary

U.S. DEPARTMENT OF COMMERCE

Dr. D. James BakerUnder Secretary of Commerce for Oceans and Atmosphere

and Administrator ofNational Oceanic and Atmospheric Administration

Terry D. GarciaAssistant Secretary of Commerce for Oceans and Atmosphere

and Deputy Administrator ofNational Oceanic and Atmospheric Administration

Dr. Nancy FosterAssistant Administrator for Ocean Services and Coastal Zone Management,

National Ocean Service

Charles W. ChallstromActing Director

National Geodetic Survey

The National Height Modernization Study was prepared forthe National Geodetic Survey by

Dewberry & Davisand Psomas & Associates

June 1998

National Height Modernization Study National Geodetic Surveyi

7$%/(�2)�&217(176

TABLE OF CONTENTS i

List of Tables v

List of Figures vi

Glossary vii

Executive Summary xi

1 --STUDY BACKGROUND 1-1

1.1 Introduction 1-1

1.2 Definitions 1-1

1.3 Need for the Study 1-5

1.4 Study Goals 1-6

1.5 Study Approach 1-6

2 --USER FORUMS 2-1

2.1 Background 2-1

2.2 Unmet Needs or Requirements 2-1

2.3 Potential GPS Applications in Height Modernization 2-2

2.4 Highest Priority Potential Applications 2-3

2.5 Federal Role in National Height Modernization 2-4

2.6 Major Messages from the User Forums 2-6

3 --USER NEED ASSESSMENTS 3-1

3.1 Application Categories 3-1

3.2 Public Safety 3-13.2.1 Police, E-911 and Fleet Vehicle Services 3-23.2.2 Disaster Preparedness and Response 3-33.2.3 Flood Mitigation 3-43.2.4 Coastal Stewardship 3-83.2.5 Seismic Monitoring 3-9

TABLE OF CONTENTS

National Height Modernization Study National Geodetic Surveyii

3.3 Transportation Management 3-123.3.1 Marine Navigation and Safety 3-133.3.2 Air Navigation and Safety 3-163.3.3 Vehicle Positioning and Safety 3-173.3.4 Train Positioning and Safety 3-17

3.4 Infrastructure Management 3-183.4.1 Water Supply and Quality 3-183.4.2 Subsidence Monitoring 3-203.4.3 Stormwater and Utility Management 3-24

3.5 Construction and Mining 3-263.5.1 Infrastructure Construction 3-263.5.2 Mining and Earth Moving 3-303.5.3 Pipeline Construction 3-32

3.6 Agriculture and Natural Resources 3-333.6.1 Precision Farming 3-343.6.2 Forestry 3-363.6.3 Recreation 3-373.6.4 Environmental Protection 3-37

3.7 USGS National Mapping Program Evaluations 3-383.7.1 User Evaluations of NAVD 88 3-383.7.2 User Evaluations of DEM Accuracy 3-393.7.3 Evaluation of DEMs from LIDAR and IFSAR Sensors 3-39

4 – CASE STUDIES 4-1

4.1 Cost Savings from GPS vs. Traditional Surveying/Leveling 4-1

4.2 Disaster Response and Recovery 4-24.2.1 Hurricane Fran (North Carolina, September 1996) 4-24.2.2 Northridge Earthquake (California, January 1994) 4-4

4.3 GPS Elevation Surveys for Infrastructure Management 4-94.3.1 Metropolitan Water District of Southern California 4-94.3.2 Greensboro (NC) Storm Water Services 4-124.3.3 Highway/Bridge Construction 4-14

4.4 Four-Dimensional Elevation Surveys (x,y,z and time) 4-164.4.1 Crustal Motion Monitoring 4-16

4.5 Subsidence Monitoring 4-174.5.1 Long Beach GPS Subsidence Network 4-174.5.2 Outside Canal and the Delta-Mendota Canal Subsidence Network 4-194.5.3 Unstable Dike Monitoring 4-214.5.4 Geodetic Control Surveys (1992 and 1998), Escondido, California 4-23

TABLE OF CONTENTS

National Geodetic Survey National Height Modernization Studyiii

4.6 GPS Control Surveys 4-264.6.1 Baltimore County, Maryland 4-264.6.2 Eastside Reservoir Project (RTK Surveys) 4-294.6.3 Geodetic Control Network, City of San Jose, California 4-344.6.4 Geodetic Control Network, Davidson County, North Carolina 4-354.6.5 GPS Control Surveys, Brunswick and Bladen Counties, N Carolina 4-36

5 -- TIME & COST COMPARISONS 5-1

5.1 Background 5-1

5.2 California Southern Project 5-35.2.1 Accuracy Comparisons 5-35.2.2 Time and Cost Data 5-4

5.3 California Northern Project 5-55.3.1 Accuracy Comparisons 5-55.3.2 Time and Cost Data 5-7

5.4 North Carolina Project 5-75.4.1 Comparison of GPS and Leveling 5-85.4.2 Time Comparison (GPS versus Leveling) 5-85.4.3 Project Statistics 5-9

5.5 Summary 5-105.5.1 Cost Comparisons - Single Baselines 5-125.5.2 Cost Comparisons - Multiple (Network) Baselines 5-135.5.3 Time Comparisons - Single Baselines 5-145.5.4 Time Comparisons - Multiple (Network) Baselines 5-15

6 -- FINDINGS 6-1

6.1 Finding #1. America needs a “reliable” and efficient means todetermine “absolute” elevations, relative only to the earth’s center. 6-1

6.2 Finding #2. Based on DGPS and CORS, America needs nationwideimplementation of a standardized vertical reference datum as the legalbasis for elevation data. 6-2

6.3 Finding #3. America needs high accuracy, high resolution DigitalElevation Models (DEMs) based on NAVD 88. 6-2

6.4 Finding #4. Based on NAVD 88, America needs a NationwideDifferential GPS (NDGPS) system for real-time accurate 3-Dpositioning; for navigation, tracking, public safety, precision farming,and construction 6-4

6.5 Finding #5. America needs cost-effective, mass-produced GPS elevationsurveys of all buildings in or near floodplains, as well as coastal areasvulnerable to hurricane tidal surges, that are based on NAVD 88. 6-6

7 -- RECOMMENDATIONS 7-1

TABLE OF CONTENTS

National Height Modernization Study National Geodetic Surveyiv

7.1 Summary 7-1

7.2 Desired Outcome/Objective 7-3

7.3 Recommendations for Implementing NAVD 88 7-47.3.1 Approach 7-47.3.2 Time Frame 7-77.3.3 Costs 7-77.3.4 Methodology 7-7

8 -- APPENDIX 8-1

8.1 National Spatial Data Infrastructure (NSDI) 8-1

8.2 National Spatial Reference System (NSRS) 8-3

8.3 Vertical Datums 8-58.3.1 Local Mean Sea Level 8-58.3.2 National Geodetic Vertical Datum of 1929 (NGVD 29) 8-68.3.3 North American Vertical Datum of 1988 (NAVD 88) 8-6

8.4 Relative vs. Absolute Accuracy 8-8

8.5 NAVD 88 Implementation 8-10

8.6 Global Positioning System (GPS) 8-10

8.7 Differential GPS (DGPS) 8-148.7.1 DGPS for Air Navigation and Safety 8-158.7.2 DGPS for Land Navigation and Vehicle Tracking 8-16

8.8 Nationwide DGPS (NDGPS) 8-18

8.9 Digital Elevation Models (DEM) 8-208.9.1 USGS DEM Types 8-218.9.2 USGS DEM Levels and Their Vertical Accuracy 8-21

8.10 Photogrammetrically Generated DEMs 8-23

8.11 LIDAR-Generated DEMs 8-24

8.12 IFSAR-Generated DEMs 8-25

TABLE OF CONTENTS

National Geodetic Survey National Height Modernization Studyv

/,67�2)�7$%/(6Table 1 Elevation Requirements for Public Safety Applications 3-2Table 2 Elevation Requirements for Transportation Management Applications 3-12Table 3 Elevation Requirements for Infrastructure Management Applications 3-18Table 4 Elevation Requirements for Construction and Mining Applications 3-26Table 5 Elevation Requirements for Agriculture & Natural Resource Applications 3-33Table 6 User Satisfaction with 7.5 Minute DEM Accuracy 3-39Table 7 User Satisfaction with 30 Minute DEM Accuracy 3-39Table 8 User Satisfaction with 1 Degree DEM Accuracy 3-39Table 9 Cost Comparison Leveling vs GPS 4-18Table 10 Eastside Reservoir Photogrammteric Mapping Costs 4-31Table 11 Eastside Reservoir GPS RTK Mapping Costs 4-31Table 12 California (South) Time/Cost Comparisons 5-4Table 13 California (North) Time/Cost Comparisons 5-7Table 14 North Carolina Geodetic Survey Time Comparisons 5-9Table 15 Statistics from Time and Cost Comparison Projects 5-10Table 16 Comparative Costs for Single Baselines 5-13Table 17 Comparative Costs for Multiple (Network) Baselines 5-14Table 18 Comparative Times for Single Baselines 5-15Table 19 Comparative Times for Multiple (Network) Baselines 5-16Table 20 High Accuracy DEM Beneficiaries 6-4Table 21 NDGPS Beneficiaries 6-5Table 22 Phase 1 - First Year Demonstration Projects 7-8Table 23 Nationwide Implementation of the Federal Base System 7-8

TABLE OF CONTENTS

National Height Modernization Study National Geodetic Surveyvi

/,67�2)�),*85(6

Figure 1 Federal Geographic Data Committee (FGDC) Structure 2-7Figure 2 Federal Geodetic Control Subcommitte (FGCS) Structure 2-8Figure 3 Increased Risk Due to Land Development in Flood Plains 3-5Figure 4 Two Houses in St. Charles County, Missouri During the Flood of ’93 3-6Figure 5 GPS Reciever in Use for Seismic Monitoring 3-11Figure 6 Subsidence in California’s Central Valley 3-21Figure 7 Annual Costs of Subsidence Mitigation 3-22Figure 8 National Distribution of Subsidence Costs 3-24Figure 9 RTK DGPS for Construction and Mining 3-31Figure 10 GPS Elevation Certificate (Post-Hurricane) 4-3Figure 11 Observations from GPS-based Earthquake Analysis 4-6Figure 12 Geodetic Damage and Restoration Following the Northridge Earthquake 4-6Figure 13 Measuring Invert Offsets for a GPS RTK Facilities’ Survey 4-12Figure 14 Subsidence Monitoring Along an Irrigation Channel 4-20Figure 15 Eastside Reservoir Project 4-30Figure 16 GPS Cost Savings (%) for Single Baselines 5-13Figure 17 GPS Cost Savings (%) for Multiple (Network) Baselines 5-14Figure 18 GPS Time Savings (%) for Single Baselines 5-15Figure 19 GPS Time Savings (%) for Multiple (Network) Baselines 5-16Figure 20 GPS Elevation Certificate (Pre-Flood) 6-6Figure 21 Vertical Datum Reference Options 8-5Figure 22 Height Measurements 8-6Figure 23 Example of Route-Dependent Elevations 8-9Figure 24 Projected Growth in GPS Users 8-11Figure 25 Operation of the FAA Wide Area Augmentation System (WAAS) 8-15Figure 26 GPS Configuration for Automated Vehicle Tracking 8-17Figure 27 Current USCG/USACE DGPS Network 8-19Figure 28 Proposed Nationwide Differential GPS (NDGPS) 8-19Figure 29 Comparison of Level 1 and Level 2 DEMs 8-23Figure 30 Comparison of Level 2 and LIDAR DEMs 8-24

TABLE OF CONTENTS

National Geodetic Survey National Height Modernization Studyvii

*/266$5<

ANA Area Navigation Approach

ASPRS American Society for Photogrammetry and Remote Sensing

AVL Automatic Vehicle Location

BARD Bay Area Regional Deformation

B/C Benefit/Cost

BFE Base Flood Elevation

BLM Bureau of Land Management

BMP Best Management Practice

CAD Computer Aided Dispatch

CADD Computer Aided Design and Drafting

CAES Computer Aided Earthmoving System

CALTRANS California Department of Transportation

CBN Cooperative Base Network

CFT Controlled Flight into Terrain

CORS Continuously Operating Reference Station

C&GS Coast & Geodetic Survey

DATIS Digital Airborne Topographic Imaging System

D&D Dewberry & Davis

DEM Digital Elevation Model

DGPS Differential Global Positioning System

DLG Digital Line Graph

DOC Department of Commerce

DOD Department of Defense

DOT Department of Transportation

DTED Digital Terrain Elevation Data

DTM Digital Terrain Model

DFIRM Digital Flood Insurance Rate Map

EOM Earth Observation Magazine

EPA Environmental Protection Agency

ERIM Environmental Research Institute of Michigan

ERM Elevation Reference Mark

FAR Federal Acquisition Regulation

FBN Federal Base Network

FBS Federal Base System

TABLE OF CONTENTS

National Height Modernization Study National Geodetic Surveyviii

FEMA Federal Emergency Management Agency

FGCS Federal Geodetic Control Subcommittee

FGDC Federal Geographic Data Committee

FHA Federal Housing Administration

FIRM Flood Insurance Rate Map

FIS Flood Insurance Study

FAA Federal Aviation Administration

FHWA Federal Highway Administration

FRA Federal Railroad Administration

GIS Geographic Information System

GPS Global Positioning System

GWEN Ground Wave Emergency Network

H&H Hydrologic and Hydraulic

HARC Houston Advanced Research Center

HARN High Accuracy Reference Network (Horizontal Control)

HAZMAT Hazardous Material

HGCSD Harris-Galveston Coastal Subsidence District

HPGN High Precision Geodetic Network

IFMRC Interagency Floodplain Management Review Committee

IFSAR Interferometric Synthetic Aperture Radar

IFSARE Interferometric Synthetic Aperture Radar for Elevation

IMU Inertial Measuring Unit

ISTEA Intermodal Surface Transportation and Efficiency Act

ITRF International Earth Rotation Service Terrestrial Reference Frame

ITS Intelligent Transportation System

JPL Jet Propulsion Laboratory

KHZ Kilohertz

LAAS Local Area Augmentation System

LF/MF Low Frequency/Medium Frequency

LIDAR LIght Detection And Ranging

MARAT Method, Accuracy, Reliability, and Application Test

MLLW Mean Lower Low Water

MSL Mean Sea Level

MWD Metropolitan Water District

NAD 27 North American Datum of 1927 (Horizontal Datum)

NAD 83 North American Datum of 1983 (Horizontal Datum)

NAPP National Aerial Photography Program

NASA National Aeronautics and Space Administration

TABLE OF CONTENTS

National Geodetic Survey National Height Modernization Studyix

NAVD 88 North American Vertical Datum of 1988

NCGS North Carolina Geodetic Survey

NDGPS Nationwide Differential Global Positioning System

NFIP National Flood Insurance Program

NGS National Geodetic Survey

NGVD 29 National Geodetic Vertical Datum of 1929

NHS National Height System

NIMA National Imagery and Mapping Agency

NOAA National Oceanic and Atmospheric Administration

NOS National Ocean Service

NPS National Park Service

NRCS Natural Resources Conservation Service

NSDI National Spatial Data Infrastructure

NSF National Science Foundation

NSRS National Spatial Reference System

NTSB National Transportation Safety Board

NYSDEC New York State Department of Environmental Conservation

OSTP Office of Science and Technology Policy

P&A Psomas & Associates

PDD Presidential Decision Directive

PGGA Permanent GPS Geodetic Array

PDC Positive Train Control

PID Permanent Identifier (ID Number)

PPS Precise Positioning Service

PTL Positive Train Location

PTS Positive Train Separation

QA/QC Quality Assurance/Quality Control

QBS Qualifications Based Selection

RMS Root Mean Square

RMSE Root Mean Square Error

RTK Real Time Kinematic

S/A Selective Availability

SCIGN Southern California Integrated GPS Network

SDWA Safe Drinking Water Act

SFHA Special Flood Hazard Area

SPS Standard Positioning Service

TEC Technical Evaluation Contractor (FEMA)

TEC Topographic Engineering Center (U.S. Army)

TABLE OF CONTENTS

National Height Modernization Study National Geodetic Surveyx

TIN Triangulated Irregular Network

UCLA University of California Los Angeles

UDN User Densification Network

URISA Urban and Regional Information Systems Association

USACE U.S. Army Corps of Engineers

USBR U.S. Bureau of Reclamation

USCG U.S. Coast Guard

USC&GS U.S. Coast & Geodetic Survey

USFS U.S. Forest Service

USGS U.S. Geological Survey

VLBI Very Long Baseline Interferometry

V-Zone Velocity Wave Action (Flood Zone)

WAAS Wide Area Augmentation System

WGS World Geodetic System

WHP Wellhead Protection Program

WHPA Wellhead Protection Area

- EXECUTIVE SUMMARY

National Geodetic Survey National Height Modernization Studyxi

(;(&87,9(�6800$5<

Introduction

This document was prepared in response to the direction contained in HouseReport 105-207 (to accompany H.R. 2267 - Departments of Commerce, Justice,and State, the Judiciary, and Related Agencies Appropriations Bill, Fiscal Year1998) that the National Geodetic Survey "conduct a National HeightModernization Study to demonstrate the effectiveness of this work in Californiaand in western North Carolina. The Committee expects the NGS to conduct thisstudy in consultation with state and local governments and the private sector."The results of the study, as abstracted by this Executive Summary, present notonly a compelling argument for the need to modernize the vertical component ofthe National Spatial Reference System, but also demonstrate how the GlobalPositioning System can be used to accomplish the modernization effort withsignificant cost savings.

A Revolution in Traditional Surveying

Though most of us are not familiar with the science of geodesy, the worldaround us is full of examples of its importance to our lives and our nation’sprosperity. The safety and efficiency of the buildings we live and work in, theroads and bridges we drive on, and the trains, airplanes, and ships that carry theproducts we use every day all depend on a universally compatible system ofgeodetic reference points that tie our nation together. Geodesy is the basicability to determine the location of a particular point in three-dimensional space,and to accurately relate it to another point. It is a fundamental necessity thatunderlies almost every facet of how the world functions today.

The mission of NOAA’s National Geodetic Survey (NGS) is to ensure that theUnited States has the consistent, high-accuracy geodetic reference framework itneeds to support these fundamental activities. Until recently, NGS has reliedon using conventional line-of-sight survey measurements to provide thatframework through a network of physical reference points accessible to usersthroughout the nation. Conventional leveling methods required crews of

The National Geodetic Survey is an office within the National Ocean Service,which is a subset of the National Oceanic and Atmospheric Administration, abureau within the United States Department of Commerce.

EXECUTIVE SUMMARY

National Height Modernization Study National Geodetic Surveyxii

geodetic surveyors to have literally walked from border to border and coast tocoast, carrying surveying equipment and taking geodetic surveyingmeasurements every hundred yards or so, to establish and maintain a nationalcoordinate system accessible to all users. In this fashion, a system of more thana million reference points was eventually built and serves today as the nation’sgeodetic reference framework.

The advent of the Global Positioning System (GPS), however, has irreversiblytransformed this landscape. Developed by the U.S. military, GPS is aconstellation of 24 satellites that transmit their signals to receivers all over theworld. GPS enables geodetic positioning to be accomplished without having tophysically see between points. Using GPS, a survey that once took days tocomplete can now be done in a few hours at a much lower cost. GPS has alsointroduced the fourth dimension of time, enabling more accurate modeling ofthe earth’s crustal motion. In addition, GPS techniques have enabled "real-time" positioning applications. As a result, GPS has not only revolutionized thetraditional civilian navigation, surveying, and mapping professions, but hasspawned numerous new applications in industrial sectors not previouslydependent on geodesy. GPS has quickly become critical to our nation'stechnological leadership and competitiveness in today's global economy.

Requirements - PhilosophyFuture GPS User Sectors - $M

(Based upon Freedonia Group Report, Cleveland, Ohio - 1997)

0

500

1000

1500

2000

2500

3000

3500

4000

1996 2001 2006

CommunicationAutomotiveGov./Inst.Civil AviationM aritimeIndustrialRecreationM ilitaryAgricultureOtherAerospace

Sales of GPS Equipment & Services

America’s Global Positioning System (GPS) already is the world’s nextUtility System. The Freedonia Group and similar organizations project ameteoric rise in GPS users over the next decade, with communications,automotive and other applications impacting virtually every American.

- EXECUTIVE SUMMARY

National Geodetic Survey National Height Modernization Studyxiii

The U.S. Department of Defense established and maintains the constellation ofGPS satellites. NGS uses GPS to provide a more accurate, underlying geodeticcoordinate system that makes possible and supports the many diverse civilianapplications of GPS technology. NGS provides the infrastructure that facilitatespublic and private civilian applications of GPS. However, GPS’s potential forinnovative applications beyond traditional uses has yet to be fully exploited.

This is because the utilization of GPS has progressed in two stages. Initially, GPSwas much more accurate in determining horizontal coordinates than verticalheights due to a number of technical factors. It is only recently that standards,specifications, and techniques have been developed, primarily by NGS incooperation with the GPS community, that enable GPS to attain the accuracylevels required for most applications utilizing height information. Unfortunately,these techniques are not yet commonly known or practiced by the private-sectorsurveying community, and require a major technology transfer effort to introducethem on a widespread basis. In addition, the existing geodetic referenceframework that supports height measurements is outdated and must bemodernized. It is unable to fully support the use of GPS to determine accurateheight measurements and therefore enable the substantial benefits possiblethrough GPS height dependent applications.

The modernized NGS satellite-based National Spatial Reference System(NSRS) is meeting that challenge by replacing the existing time-consuming,labor-intensive framework with a significantly smaller network designed tosupport and enhance the technological advantages of GPS. The modernizedNSRS is easier to maintain and 10 to 100 times more accurate in the horizontaldimension than the previous system. NSRS maximizes the potential of GPS byenabling GPS methods to determine height measurements to the accuraciesrequired for their respective applications, as well as bridging the gap betweenGPS and pre-existing reference systems.

In many respects NSRS can also be thought of as the foundation for theNational Spatial Data Infrastructure (NSDI), a critical component of the"information superhighway". NSDI facilitates data sharing by organizing andproviding a structure of relationships between producers and users of spatialdata and thus ensures consistent and reliable means to share spatial data.

NGS has recently completed the major portion of the horizontal component ofNSRS by leveraging appropriated funding through other Federal, state, and localgovernment entities. However, the vertical component of NSRS, the NationalHeight System (NHS), presents a bigger challenge to modernize. Urbanizationand construction have destroyed many of the original reference survey points

EXECUTIVE SUMMARY

National Height Modernization Study National Geodetic Surveyxiv

that NHS is founded upon, and numerous others have succumbed to the effectsof subsidence and seismic activity. As a result, the system is unreliable in manyareas, and nonexistent in others. Until recently, only conventional verticalsurveying methods could be used to implement NHS due to accuracyrequirements. Fortunately, the recent development of NGS technical guidelinesand techniques now offers the prospect that GPS can be used to accomplish themodernization effort at much lower cost.

This study assesses the need(s) for, and benefits to be derived from, a modernizedNSRS/National Height System, and in turn the many existing and potential GPStechnology applications it will support. The study also evaluates the technical,financial, legal, and economic aspects of using GPS technology to modernizeNHS. The study presents findings, and makes recommendations for theirimplementation.

Study Scope and Methodology

NGS established the following major goals for the study:

• Identify and document user requirements for height data, including thoserequirements utilizing both vertical and horizontal data.

• Identify and document major users and applications of the NationalHeight System and GPS-derived height data.

• Identify and recommend the best, most cost-effective actions that meetthe documented user requirements, taking into consideration alltechnologies available.

• Evaluate the estimated costs to implement the recommended actions, andtheir benefits to the nation.

NGS, with the assistance of private sector consultants, conducted the study tomeet these goals by utilizing three main mechanisms:

• Utilization of User Forums and other outreach means to obtain theinsights and recommendations of state and local governments, theprivate sector, and other interested parties to:

- identify outstanding needs and requirements with respect toheight information and technology,

- EXECUTIVE SUMMARY

National Geodetic Survey National Height Modernization Studyxv

- assess current and potential applications of the NationalHeight System and GPS height data, and

- document their value to the nation.

• Determination of the relative costs and benefits of GPS technologyversus conventional methods for height surveys by evaluating 16 casestudies. This step involved having users conduct post-analyses of pre-existing, ad-hoc GPS height survey projects in California, NorthCarolina, and other selected locations.

• Determination of the relative costs and benefits of GPS technologyversus conventional methods for height surveys by contracting forsurveys to be conducted utilizing both methods over identical projectareas.

NGS consultants were responsible for the collection, assessment,analysis and reporting of the information gathered by the above steps forthe study.

Study Results

The following summarizes the information gathered through the study’s UserForums and cost-benefit evaluations.

User Forums

The most common themes identified by users with regard to unmet needs andrequirements for height information, as grouped into broad categories, were:

• The need for a reliable, cost-effective, standardized, legally establishednational vertical reference datum and the infrastructure (NSRS) throughwhich to access and utilize it.

• The capability to easily inter-relate the many vertical datums currently inexistence, but particularly with respect to a standardized national verticalreference datum.

• The need for national technical standards and guidelines for using GPS todetermine heights.

• The need for improvements to NSRS, particularly the National HeightSystem component, to improve access throughout the nation, includes:

EXECUTIVE SUMMARY

National Height Modernization Study National Geodetic Surveyxvi

- implementation of the National Height System network ofsurvey points at a nominal spacing of 10 kilometers by 10kilometers,

- densification of the GPS Continuously Operating ReferenceStation network,

- improvements to the geoid models required to relate GPSdetermined heights to those determined through conventionalsystems, and

- improved infrastructure support for real-time GPSpositioning.

Existing and potential GPS height application themes that were frequentlyidentified by the user community were grouped into five major summarycategories:

• Public Safety

• Transportation Management

• Infrastructure Management

• Construction and Mining

• Agriculture and Natural Resources.

The study examined each of these categories and provides in depth assessments oftheir existing applications, potential applications, and benefits. The study alsoexamined the support provided, or potential benefits that would be realized, forthese applications from an accessible National Height System. A cross section ofjust some of the many benefits include:

• Improved coastal and harbor navigation, enabling safer and more cost-effective shipment of goods.

• More efficient fertilizer and pesticide application, reducing runoff ofchemicals that result in water pollution, and enhancing economiccompetitiveness through lower costs and higher crop yields.

• Accurate digital elevation models, enabling better floodplain analysisand determination of flood insurance needs.

- EXECUTIVE SUMMARY

National Geodetic Survey National Height Modernization Studyxvii

• Enhanced airline and aircraft safety through GPS-controlled approachand landing systems.

• More accurate models of storm surges, coastal erosion, and trajectoriesof oil and chemical spills that enable better response to these hazards.

• Improved understanding of tectonic movement.

• Better management of natural resources through the use of reliablegeographic information systems (GIS).

• Support for the modernization of America’s transportation infrastructureand the mission and goals of the Intermodal Surface Transportation andEfficiency Act.

• Accurate and consistent elevation data for building environmentallysustainable cross-border projects with Canada and Mexico.

The study gathered information and assessed the financial benefits, either realizedor projected, of the applications identified within the summary categories. Asseen on the next page, the potential for financial benefits was found to bestaggering, even based on conservative estimates.

EXECUTIVE SUMMARY

National Height Modernization Study National Geodetic Surveyxviii

Areas benefitingfrom amodernizedNational HeightSystem

Estimated Valueto Constituents

Explanation of Benefits

Nationwide Terrain $33.5 million • Replace less-accurate Level 1DEMs that cost USGSapproximately $33.5 Million

• Enable rapid generation ofcontours for USGS maps andGISs nationwide

• Enable 3-D modeling byUSACE, FHA, FRA, FAA,EPA, USFS, etc.

Nationwide Watersheds $100 million • Automated hydrologic modelingby NWS and FEMA to predictlocations/ volumes of peakwater concentrations

Special Flood HazardAreas (SFHAs)

$225+ million • Automated hydraulic modelingby FEMA to determine depthand extent of flood waters

• Determination of flood risks andinsurance rates

Coastal Erosion Zones $11.25+ million • Accurate determination ofcoastal erosion rates

• Determination of insurance ratesUrban Areas $500 million • Urban planning

• Intelligent TransportationSystem (ITS) planning

• Elevation layer in GIS database• Stormwater management

Farm Lands $1.7 billion • Precision farming for plannedapplication of water, fertilizer,etc.

• Control of unwanted run-off andstream contamination

Maritime Navigation andSafety

$9.6 billion • Positioning of dredges• Positioning of cargo ships

Surveying Industry Not estimated • Vastly improved surveyprocedures

Totals $12+ billion •

Table of estimated benefits from a modernized National Height System summarized fromseveral of the Study tables

- EXECUTIVE SUMMARY

National Geodetic Survey National Height Modernization Studyxix

Case Studies

This step summarizes detailed reports submitted to NGS from users in the fieldproviding data on both the cost benefits and other lessons learned through the useof GPS techniques over conventional surveying methods. A variety of differenttypes of surveying projects were selected to provide a good cross section ofapplications. Most of the case study GPS surveys were conducted prior to NGS’publication of its GPS height survey guidelines. It was found that:

• The cost savings realized from using GPS versus conventional surveyingmethods ranged from 25 percent to more than 90 percent, dependingupon the type of survey conducted. The large range of cost savings wasprimarily due to the type of project being conducted. However, thesecost savings should be qualified as follows. It is expected that had themore rigorous NGS guidelines been followed, overall cost savings mayhave been lower, but quality assurance and data reliability would havebeen much higher. This potential reduction, however, would be offsetby the greater efficiencies gained through adherence to established NGSguidelines.

• Cost savings were greatest when large distances (> 4km) existedbetween survey points, and there was no requirement to establish heightsat intermediate points. Cost savings diminish, or are negligible, whendistances are small (< 2 km) between survey points.

• When NGS guidelines were followed, GPS-derived heights wereaccurately determined within the range needed by most users, e.g., 2centimeters (3/4 inch) for their applications.

• A common, reliable vertical datum (e.g., NAVD 88) that is easilyaccessible through an existing infrastructure (NSRS/NHS) is essentialfor the use of GPS-determined heights to be possible and effective.

• Utilizing GPS to conduct a survey has an added benefit of providinghorizontal coordinates in addition to height information. Two separatesurveys would have to be run using conventional surveying methods toaccomplish the same results at obviously much higher cost.

• GPS is particularly effective when geodetic control must be quickly re-established in disaster areas where the local geodetic infrastructure hasbeen largely destroyed, when a project involves large areal coverage, orwhen difficult, rugged terrain lies between survey points.

EXECUTIVE SUMMARY

National Height Modernization Study National Geodetic Surveyxx

• A combination of GPS and conventional surveying methods often makesthe most effective use of both.

• The number of primary control points required can often be significantlyreduced using GPS techniques.

Cost Comparison Surveys

Contracts were let in North Carolina and California to survey the same surveypoints (within each contract) by first using one method, and then repeated usingthe other method (GPS technology and conventional surveying), to determinethe relative differences in costs, accuracies, and other factors under controlledconditions, including being run using NGS guidelines. It was found that:

• The contract surveys confirmed the significant cost savings anddesired accuracy level results found in the case studies, as well asthe other additional benefits identified by the case studies. Thecost savings are discussed in more detail below.

• In general, conventional surveying methods are more accuratethan GPS methods for height determination over relatively shortdistances. However, at the 10 kilometer spacing recommendedfor the National Height System, GPS accuracies are comparablewith those attained by conventional surveying for mostapplications.

Variable Cost Savings from GPS

• Post Hurricane Elevation Surveys 90%

• Post Earthquake Elevation Surveys 66%

• Water District Elevation Surveys 75%

• Crustal Motion Monitoring 99%

• Subsidence Monitoring 45 - 75%

• GPS RTK Construction Surveys 26 - 71%

• County- and City-wide 3-D Control Surveys 26 - 80%

• Topographic Mapping for Reservoir Construction 71%

- EXECUTIVE SUMMARY

National Geodetic Survey National Height Modernization Studyxxi

• One of the surveys highlighted an additional benefit of utilizingGPS to determine heights. A difference in the results obtained bythe two methods identified the need to refine the geoid model(used to convert GPS-determined heights to those determined byconventional methods) for the local area. This benefit will berealized for other areas if GPS methods are used to implement theNational Height System.

Comparison of GPS versus conventional surveying costs are best discussed interms of survey points connected and the distance in between those points.Within reason, the distance between survey points is not a cost factor for GPS, butis a major factor for conventional surveying methods.

In general, survey points that are relatively close together (< 3 km) can havetheir heights determined for less cost by using conventional methods over GPSmethods (although GPS also provides horizontal coordinates at no additionalcost). However, once this minimum distance is exceeded, cost savings fromusing GPS methods rapidly accrue. Because GPS costs essentially remainconstant for two survey points (within reason), the greater the distance betweensurvey points, the larger is the savings. The following tables are based on ananalysis of the results of the contract surveys.

The first table shows cost comparisons assuming a survey conducted betweentwo points and an increasing distance. GPS methods result in cost savingssomewhere between 2 and 3 kilometers.

For two points connected by a single baseline with lengths between 1 and 10kilometers, the comparative costs between leveling and GPS are shown below.

SingleBaseline

LevelingCosts

GPSCosts

GPSSavings

1 Km $680 $1,620 -138%

2 Km $1,360 $1,620 -19%

3 Km $2,040 $1,620 21%

4 Km $2,720 $1,620 40%

5 Km $3,400 $1,620 52%

10 Km $6,800 $1,620 76%

Comparative Costs for Single Baselines

BASELINE

EXECUTIVE SUMMARY

National Height Modernization Study National Geodetic Surveyxxii

The second table shows cost comparisons for expanding a surveynetwork, based on a four-point box, one box at a time. This example isconsistent with how the National Height System would be implementedat the recommended 10-kilometer spacing. Note that using GPSmethods result in almost a 90 percent cost savings over conventionalsurveying methods.

For an expanding network with four new points at a time forming multiplebaselines, with each baseline length between 1 and 10 kilometers, thecomparative costs between leveling and GPS are shown below:

SingleBaselineLength

TotalLevelingLength

LevelingCosts

GPSCosts

(4 i t )

GPSSavings

1 Km 4 Km $2,720 $3,240 -19%

2 Km 8 Km $5,440 $3,240 40%

3 Km 12 Km $8,160 $3,240 60%

4 Km 16 Km $10,880 $3,240 70%

5 Km 20 Km $13,600 $3,240 76%

10 Km 40 Km $27,200 $3,240 88%

Comparative Costs for Multiple (Network) Baselines

BASELINE

4-POINT BOX

- EXECUTIVE SUMMARY

National Geodetic Survey National Height Modernization Studyxxiii

Major Findings of the Study

Five major findings were identified from the user forums.

1. The Nation needs a reliable and efficient means to determine "absolute”heights. This can best be accomplished through the use of GPStechnology in conjunction with NGS technical guidelines and anaccessible National Height System.

2. The Nation needs a reliable, cost-effective, standardized, legallyestablished national vertical reference datum (NAVD 88) and theinfrastructure (NHS) through which to access and utilize it.

3. The Nation needs high accuracy digital elevation models based onNAVD 88 as supported by the National Height System.

4. The Nation needs a nationwide differential GPS system to supportaccurate real-time three-dimensional applications.

5. The Nation needs cost-effective elevation surveys of floodplains andcoastal areas vulnerable to flooding that are based on NAVD 88. GPS isthe best technology to accomplish this.

Study Recommendations

The study clearly identified the need for a National Height System(NHS) and that GPS technology, conducted in accordance with NGSguidelines, is the most cost-effective method to implement it.

This study provides recommendations on how to best accomplish thisand satisfy the first two Major Findings, as described below. The lastthree Major Findings are in the process of being implemented by otherFederal agencies and their partners, but would be greatly enhanced by anaccessible NHS.

The study recommends a two-phase approach to modernize and sustainNHS throughout the conterminous United States. Phase 1, described indetail in the report, would focus on the survey and scientific workneeded to establish the basic framework of NHS, while Phase 2,addressed in the report but in less detail, would ensure the system'scontinued sustainability.

EXECUTIVE SUMMARY

National Height Modernization Study National Geodetic Surveyxxiv

Phase 1 - Establish the System’s Framework

Phase 1 would involve a cooperative effort among the private sector andFederal and state agencies to establish NHS’ basic framework ofreference survey points, a network of 55,000 bench marks with 10 km by10 km nominal spacing called the Federal Base System, with theirheights accurately determined by GPS. Phase 1 would be accomplishedover a 5-year span, utilizing a state-by-state approach.

The first step of Phase 1 would be to conduct two prime demonstrationprojects in California and North Carolina during the first year. Thesestates are subject to extreme seismic activity, subsidence, riverine andcoastal flooding, coastal erosion, and accelerating development. They areideal for fully refining the GPS techniques and guidelines needed to mosteffectively implement NHS, and conduct the training and transference ofthis technical expertise to the state and private partners needed to developthe large institutional capacity required to implement NHS on a nationalbasis. This expanded, constantly growing technical capacity will, as thenext step, enable NHS to be implemented in a larger number of states eachyear as the next step.

During all steps of Phase 1, the private sector would perform thegeodetic survey projects under NGS oversight and technical guidelines.The private sector projects are the most resource intensive aspect of theactivities needed to establish the framework. In general, these projectsconsist of activities such as:

• identifying existing, or establishing new, survey points suitablefor incorporation into NHS;

• collecting, processing, and adjusting GPS and leveling dataconnecting these survey points into NHS;

• preparing appropriate documents and reports; and

• submitting results to NGS for quality control and assimilationinto NHS.

NGS oversight and documentation responsibilities would include thefollowing:

- EXECUTIVE SUMMARY

National Geodetic Survey National Height Modernization Studyxxv

• managing the implementation effort, determining priority areas,selecting qualified private-sector contractors, managing contracts,and providing technical supervision;

• analyzing, documenting, and publishing the accuracy of the GPS-derived heights and accuracies obtained by the projects; and

• performing technology transfer activities to promoteunderstanding and implementation of the National Height Systemincluding: publishing documents, conducting seminars andworkshops, and developing training seminars for technicalpersonnel to become instructors.

The following tables estimate and compare the costs of usingconventional surveying methods versus GPS technologies to completePhase 1 of the National Height System modernization. The higher costsshown for conventional surveying techniques are the direct result of thesignificantly higher time and labor demands. It should be noted that thisstudy only estimates the overall costs to implement the National HeightSystem and makes no statement or implication about the source(s) ofthese funds.

Activity Estimated Cost of UsingConventional Surveying

Technologies

Estimated Cost ofUsing GPS

Technologies

State California NorthCarolina

California NorthCarolina

Subtotal $41,200,000 $20,040,000 $4,600,000 $2,380,000

TOTAL $61,240,000 $6,980,000

Comparative Costs of Implementing Phase 1 NHSDemonstration Projects During the First Year

EXECUTIVE SUMMARY

National Height Modernization Study National Geodetic Surveyxxvi

Activity Estimated Costs ofUsing Conventional

Surveying Technologies

Estimated Costs ofUsing GPS

Technologies

TOTAL $596,000,000 $66,000,000

Phase 2 - Ensure the System’s Sustainability

The need for Phase 2 was addressed in the full report, but was not fleshed out ingreat detail. State, local, and private sector partners would be responsible formaintaining NHS, and potentially expanding it into urban or other areasrequiring a denser network for greater accessibility. This would be similar tothe systems in place to maintain the horizontal component of NSRS. The studyparticipants agreed that maintaining the system’s sustainability wouldnecessitate NGS’ continued provision of technical leadership, programoversight, and security.

Comparative Costs of Implementing the National HeightSystem over 5 years

National Height Modernization Study National Geodetic Survey1-1

� ��678'<�%$&.*5281'

1.1 Introduction

This report was prepared in response to the direction contained in HouseReport 105-207 (to accompany H.R. 2267 – Departments of Commerce,Justice, and State, the Judiciary, and Related Agencies AppropriationsBill, Fiscal Year 1998) that the National Geodetic Survey (NGS)“conduct a National Height Modernization Study to demonstrate theeffectiveness of this work in California and in western North Carolina.The Committee expects the NGS to conduct this study in consultationwith state and local governments and the private sector.” The results ofthe study not only present a compelling argument for the need tomodernize the National Height System (NHS) – the vertical componentof the National Spatial Reference System (NSRS) -- but also show howto use the Global Positioning System (GPS) to accomplish heightmodernization with tremendous cost savings. GPS and Differential GPS(DGPS), both vital to height modernization, are explained in detail insections 8.6 and 8.7 of the Appendix.

1.2 Definitions

National Spatial ReferenceSystem (NSRS). The NSRS isthat portion of the NationalSpatial Data Infrastructure(NSDI) that provides accuracy to2- or 3-dimensional geospatialdata. In many respects, the NSRSis the foundation of the NSDI,both of which are explained indetail in sections 8.1 and 8.2 ofthe Appendix.

Heights vs. Elevations.Throughout the report, the terms“height” and “elevation” areused interchangeably andgenerically to avoid confusioncaused by the use of precisely-defined technical terms, e.g.,orthometric heights, ellipsoidheights, dynamic heights, andgeoid heights.

-- STUDY BACKGROUND


National Height System (NHS). The NHS is the vertical component ofthe NSRS, i.e., that portion of the NSRS that determines elevations andelevation accuracies.

For centuries, America’s heightsystems have relied upontraditional leveling, also knownas differential leveling orconventional leveling. Levelingis based on a series of line-of-sight measurements of elevationdifferences, measured inlandfrom surveyed benchmarks withelevations referenced to mean sealevel (msl). The NationalGeodetic Vertical Datum of 1929(NGVD 29), for example,assumed that 21 tidal stations inthe United States represented thesame (zero) elevation above msl.However, modern surveys provethat msl in Boston, for example,is a different elevation than msl atother tidal stations along theAtlantic, Gulf, and Pacific coasts.This is one of numerous technicalreasons why NGVD 29 elevationsneed to be corrected by offsets,varying up to five feet, forconversion to NAVD 88elevations.

Leveling uses optical survey instruments that have been “leveled”relative to the local direction of gravity. However, the local direction ofgravity varies as a result of mass excesses or deficiencies in the earth,causing the geoid (the equipotential surface with gravity equal to that atmsl) to undulate up and down at variable distances above or below thereference ellipsoid. The reference ellipsoid is used for determination ofhorizontal and vertical coordinates of points on or near the earth’ssurface.

Leveling in this report issynonymous with conventional,traditional, or differential levelingperformed by surveyors who first“level” their optical surveyinstruments to obtain line-of-sightperpendicular to the local directionof gravity. Precise or geodeticleveling refers to the moreaccurate forms of leveling. Suchsurveys are primarily dependent onthe rules of gravity.

GPS surveys in this report areusually synonymous with GPSvertical, height, or elevationsurveys, GPS 3-D positioning, orGPS leveling. GPS surveys areessentially independent of thelocal direction of gravity and donot require line of sight betweenpoints surveyed. GPS horizontalsurveys are described as such, oras GPS 2-D surveys. GPS surveysare primarily dependent on therules of geometry.

-- STUDY BACKGROUND

National Geodetic Survey National Height Modernization Study1-3

Elevations derived from leveling have accuracies relative to numerousfactors explained in the Appendix, whereas elevations derived from GPSvertical surveys can essentially have absolute accuracy, relative only tothe earth’s center. The accuracy of elevations from leveling is typicallyspecified in relative terms, e.g., 0.4-mm times the square root of thelength of the level loop surveyed (in kilometers). Alternatively, theaccuracy of elevations from GPS surveys are typically specified inabsolute terms, e.g., ±2-cm at the 95% confidence level.

The modernized NHS corrects major deficiencies of the past andpermits elevations to be determined with accuracies that approach“absolute.” The modernized NHS consists of the following:

• 3-D CORS Stations (zero errors) comprise a nationwide network ofhundreds of 3-D Continuously Operating Reference Stations (CORS)surveyed by NGS or other Federal or state agencies (using NGS’most rigorous specifications). With continuous, ongoing surveys,relative to the full constellation of GPS satellites, CORS stations aresurveyed so well that their elevations are assumed to have zero errorsrelative to the center of the earth. Although all surveys are relative,elevation surveys relative to CORS approach the pure definition ofabsolute accuracy. Some CORS stations provide data for highaccuracy GPS post-processing of elevation data, while other CORSstations are radio beacon sites that instantaneously transmitdifferential GPS corrections for users with GPS real-time kinematic(RTK) applications. See the Appendix for additional details.

• 3-D HARN Stations comprise a network of thousands of 3-D HighAccuracy Reference Network (HARN) stations surveyed by NGS orother Federal or state agencies. These are well-documented,permanent survey monuments that are available for public use foreither GPS or traditional elevation surveys.

• “Blue Booked” survey monuments comprise a network of hundredsof thousands of 2-D and 3-D survey monuments that have beensurveyed over the years and maintained by NGS in the NSRS.Survey Data Sheets for each survey monument are easily available tothe public on CD-ROM or via the Internet. When used as a noun,blue book refers to a 3-volume set of publications (with blue covers)entitled: “Input Formats and Specifications of the National GeodeticSurvey Data Base.” It is a user’s guide for preparing and submittinggeodetic data for incorporation into NGS’ data base. Survey data

-- STUDY BACKGROUND


that are entered into NGS’ data base become part of the NSRS. Theguide comprises three volumes: volume I covers horizontal geodeticdata, volume II covers vertical geodetic data, and volume III coversgravity data. When used as a verb, blue booking refers to the processof submitting survey data to NGS consistent with rigorous blue bookprocedures for preparing and submitting geodetic data for use byothers.

• The Geoid Model (currently Geoid 96) is NGS’ mathematical modelthat provides the geoid height, i.e., the distance of the geoid above orbelow the ellipsoid. For any given latitude and longitude, the geoidheight is necessary for conversion between orthometric heights (fromleveling) and ellipsoid heights (from GPS).

• The Vertical Datum is the basis of reference for all elevationsurveys. Vertical datums are explained in detail in the Appendix.The North American Datum of 1983 (NAD 83) replaced the NorthAmerican Datum of 1927 (NAD 27) years ago as the modernizedhorizontal datum within the NSRS. However, the North AmericanVertical Datum of 1988 (NAVD 88) has never been implementednationwide as the modernized replacement for the National GeodeticVertical Datum of 1929 (NGVD 29) in spite of its officialdesignation by the Federal Geodetic Control Subcommittee (FGCS)of the Federal Geographic Data Committee (FGDC) in 1993.1

Digital Elevation Models (DEMs) are digital files of ground points for whichlatitudes, longitudes, and heights are known. According to the AmericanSociety for Photogrammetry and Remote Sensing (ASPRS)2, DEMs areuniformly spaced points (grid or lattice) such as USGS DEMs with standardpoint spacing of 30m x 30m or 100m x 100m. But the term “DEM” is also usedin a generic sense to include digital elevation data that are non-uniformlyspaced, such as Digital Terrain Model (DTM) and Triangulated IrregularNetwork (TIN) data produced by a variety of means. DEMs are often used as

1 Federal Register Notice, Vol. 58, No. 12, June 24, 1993, designated NAVD 88 as theofficial replacement of the older NGVD 29 vertical datum

2 Maune, David, 1996, “Introduction to Digital Elevation Models (DEM), DigitalPhotogrammetry, An Addendum to the Manual of Photogrammetry, AmericanSociety for Photogrammetry and Remote Sensing (ASPRS), pp. 131-134.

-- STUDY BACKGROUND


the base for topographic and hydrographic mapping and as the heightcomponent of Geographic Information Systems (GIS).

1.3 Need for the Study

NAVD 88 is an elevation reference system designed to be integrated intoa seamless network of horizontal and vertical reference points, gravitydata, GPS satellites, and tracking stations. It supports such diversifieduses as:

• Precise navigation and aircraft landing systems

• Floodplain management and the National Flood InsuranceProgram

• Highway and railroad infrastructure systems

• Intelligent vehicle highway systems

• Earthquake, volcanic, and subsidence research programs

• Disaster preparedness and relief efforts

• Water transportation infrastructure

• Precision agriculture

The change from NGVD 29 to NAVD 88 was necessary to providefor three-dimensional positioning on a global level, such as isafforded by GPS, and to remove distortions in the older verticaldatum as described by Zilkoski et. al.3. In the conterminous UnitedStates, the height differences between NGVD 29 and NAVD 88 varybetween –40-cm (-16 inches) and +150-cm (+59 inches).

Implementation of NAVD 88 means developing the network throughrecompilation of existing data and execution of new surveys and studiesto bring the horizontal, vertical, and gravity control networks togetherinto a unified system, joined and maintained by GPS. NGS conductssuch surveys under the authority of 33 U.S.C. 883a et seq. and the Officeof Management and Budget (OMB) Circular A-16 (Oct. 19, 1990).

3 Zilkoski, David B., Richards, John H., and Young, Gary M., 1992, “Results of theGeneral Adjustment of the North American Vertical Datum of 1988,” Surveying andLand Information Systems, Vol. 52, No. 3, pp. 132-137.

-- STUDY BACKGROUND


.

1.4 Study Goals

The goals of the National Height Modernization Study are to:

• Identify and document user requirements for height data,including those requirements utilizing both vertical andhorizontal data.

• Identify and document major users and applications of theNational Height System and GPS-derived height data.

• Identify and recommend the best, most cost-effective actionsthat meet the documented user requirements, taking intoconsideration all technologies available.

• Evaluate the estimated costs to implement the recommendedactions, and their benefits to the nation.

1.5 Study Approach

The study included the following:

• User forums held in California and North Carolina to determine whothe elevation “users” and “providers” are; to determine user needsassessments by application categories; and to define the role of NGS insatisfying those requirements

• Evaluation of technological opportunities to satisfy these needs

• Compilation of GPS case study “lessons learned” from leading surveyfirms knowledgeable with regard to NGVD 29 and NAVD 88

• Cost comparisons between GPS and leveling from controlled testsurveys

• Analysis of options and development of conclusions andrecommendations

-- STUDY BACKGROUND


The results are a comprehensive analysis of elevation needs, andtechnical proposals for satisfying those needs, whether provided byNGS or others.


� ��86(5�)25806

2.1 Background

This chapter summarizes the results of user forums, sponsored by NGSin Ontario, California, and Raleigh, North Carolina on January 26-27,1998 and February 4-5, 1998, respectively. The purpose was to provideinformation regarding the study and to obtain insights and suggestionsfrom individuals in the community of spatial data users, includingsurveyors, engineers, mappers, shippers, and marine pilots, forcontribution to the study and aid in evaluating the technical, financial,legal, and economic aspects of modernizing National Height Systemtechnology.

The forum participants represented a wide range of spatial data usercommunities. The forum format included an overview by NGS of thecurrent efforts, and panels of GPS advocates whose objective was tochallenge participants and provoke thinking about the potential formodernizing the National Height System. The participants providedtheir individual feedback and recommendations on the following:

• Existing unmet user requirements

• Potential applications of GPS

• An indication of priorities and the proposed Federal role in meetingunmet requirements.

The following sections are a synthesis of the inputs from the two userforums.

2.2 Unmet Needs or Requirements

In an introductory ‘brainstorm” session, the interactive discussion groupspresented a wide range of unmet needs and requirements for height data.The most common themes reflected needs for the following:

-- USER FORUMS


• North American Vertical Datum of 1988 (NAVD 88) densification andimprovement

• Relating existing vertical datums, including tidal datums, to oneanother

• An improved geoid model, necessary for adjusting GPS-derivedellipsoid heights to leveling-derived orthometric heights

• Continuously Operating Reference Station (CORS) densification andimprovement

• National standards and guidelines for GPS-derived orthometric heights

• Improvements to the National Spatial Reference System (NSRS)

• Infrastructure for real-time 3-D positioning with GPS

2.3 Potential GPS Applications in Height Modernization

Subsequently, the interactive groups discussed additional GPSapplications in height modernization not offered by panelists or NGS.These ranged from using GPS for mapping, navigation, safety, andscientific applications, to using it for land-based private and commercialactivities such as ski resort and golf course facilities management,fishing, agriculture, land fill height monitoring, forestry, and snow plowmanagement. Major GPS application themes that frequently arose fromvarious participants included:

• Land Transportation and Safety: real-time positioning of vehicles;transportation of cars, trains, trucks, buses, emergency vehicles, andheavy equipment; management of traffic congestion

• Marine Navigation and Safety: controlled dredging and hydrographicsurveys; provision of accurate, real-time information on under-keelclearance; collision and grounding avoidance; improvements inshipping cost-effectiveness

• Air Navigation and Safety: avoidance of mid-air collisions andcontrolled flight into terrain (CFT); zero-visibility landings; flightsimulations; airport and airspace elevations and obstructions; security

• Infrastructure Management: modeling and monitoring the nation’sinfrastructure; monitoring of subsidence and crustal motion, ground

-- USER FORUMS


and subsurface water levels; hydrologic and hydraulic modeling; 3-Dlocation of utilities – above/below ground; mining and construction.

• Precision Farming: water delivery and irrigation control; drainage;fertilizer/pesticide delivery and application; monitoring plant growthsand yields; vegetation and turf management; erosion and sedimentationmodeling, assessment, and management; and mitigation of non-pointsource pollution.

• Environmental Protection: environmental assessments; effects ofocean rise; hazardous incident monitoring; natural resourcemanagement

• Flood Mitigation: proactive floodplain management, flood hazards andrisk assessments, floodproofing initiatives, strengthening the NationalFlood Insurance Program (NFIP)

• Public Safety: Hazard monitoring (seismic and dam deformation);emergency response; snow removal management

2.4 Highest Priority Potential Applications

Next, the user forum participants focused on applications believed tohave the highest potential for GPS. The themes most frequently namedas highest potential were:

(1) Need for vertical accuracy and consistency among GIS databases,especially as used for:

• Floodplain management and hazard mitigation

• Public works/infrastructure management

• Subsidence/crustal motion monitoring

(2) Need for real-time GPS applications, especially for the following:

• Air and marine navigation

• Land-based vehicles, including construction equipment


• Emergency response

-- USER FORUMS


2.5 Federal Role in NationalHeight Modernization

Finally, the user forum groupsprovided feedback on the Federalrole in National HeightModernization. The participantsdescribed the desired Federal role asfollows:

(1) In setting specifications,guidelines, and/or standardsfor elevation surveys anddata, NGS’ role should be:

• Developing guidelines foruse of GPS technology toachieve desired accuracy.

• Developing programs foradoption by state, regional,and local governments tomanage geodetic controlissues, and establish auniform format forexchange among geodeticor GIS software andhardware vendors.

• Being more involved in influencing international standards,specifications, and guidelines.

(2) In establishing strong national leadership, NGS’ role shouldinclude:

• Coordination of the evolution and nationwide implementation ofthe modernized National Height System (NHS)

• Continued research in new technology (in-house and throughgrants)

• Work with GPS manufacturers for standardization of GPS data

Specifications orGuidelines?

There is no significantdifference in content betweenspecifications and guidelines.They each outline a set of stepsor procedures that result inthe achievement of specificgoals or standards. Thedifference is in their perceivedlevel of authority.Specifications are seen asrigid, while guidelines aremore flexible, a set ofprocedures recommended forreaching a specific goal, butrecognizing that they are notnecessarily the only path tothat goal. In the case of anational height standard, GPStechniques can be used inattaining a certain level ofpositioning accuracy.

-- USER FORUMS


• Serving as a catalyst to establish the number of base stationsnecessary for 1 cm accuracy, the highest accuracy requirementspecified by the users

• Exploring alternative reference systems, e.g., the InternationalEarth Rotation Service Terrestrial Reference Frame (ITRF)

• Ensuring that all projects using Federal funds are “blue booked”for inclusion in the National Spatial Reference System (NSRS)

• Transferring new technology

(3) For maintaining the database at a Cooperative Base Network(CBN) level, NGS’ role should include:

• Establishing and maintaining the physical framework

• Working with partners at the local level to maintain the network

• Maximizing the use of local surveyors to accomplish surveys

• Maintaining and distributing data; reviewing and ensuring thequality of NSRS

• Enhancing existing orthometric height system standards aroundNAVD 88

• Coordinating and overseeing orthometric heights on HighAccuracy Reference Network (HARN) stations and tidal andlake bench marks

• Developing and distributing an improved geoid model, andmaintaining the capability to establish the highest order surveynetwork

• Coordinating expansion of CORS and beacon stations tocentimeter accuracy nationwide

(4) For verifying the consistency of data, NGS’ Federal role shouldinclude documenting data quality and metadata, and coordinatingdata distribution by others.

(5) For working with the user community, NGS’ role should include:

• Strengthening the role of the State Geodetic Advisors

• Transferring new technology and sponsoring trainer training

• Enhancing educational outreach through new technology, e.g.,Internet, video conferencing

-- USER FORUMS


• Increasing staff capabilities to test and evaluate new equipment

• Providing reports to users

• Ensuring the use of user friendly software for NSRS data

• Providing easy access data retrieval, e.g., via the Internet

(6) Serving as a data and information clearinghouse for the NSRS.

(7) Advocating and serving as the key GPS interface with theAmerican Congress on Surveying and Mapping (ACSM), theAmerican Society for Photogrammetry and Remote Sensing(ASPRS), the Urban and Regional Information SystemsAssociation (URISA), etc.

2.6 Major Messages from the User Forums

There was a high level of support for modernizing the NHS. Participantsparticularly highlighted key roles for GPS in land, air, and marinetransportation; infrastructure management; agriculture; flood mitigation;and emergency management.

Consistent with theresponsibilities of the FederalGeographic Data Committee(FGDC), shown in Figure 1, andthe Federal Geodetic ControlSubcommittee (FGCS), shown inFigure 2, the Federal role infuture use of GPS for heightmodernization was supported.This included greater Federalinvestment and coordination withone agency in the lead role forGPS development andapplication; the need for the U.S.to maintain world leadership inGPS standards and promotenational and internationalconsistency; and reduceddependency on leveling.

Unmet needs focused on:

(1) Standards and guidelinesfor GPS surveys

(2) Conversion/updating fromNGVD 29 to NAVD 88

(3) Increased densification ofvertical control

(4) Centralized source of GPSinformation

(5) Improved data systemdistribution

(6) An improved geoid model

(7) Combining of traditionaland GPS leveling

(8) Infrastructure to supportreal-time 3-D positioning.

-- USER FORUMS


Highlights of comments regarding users and partners included the ideasthat future use of GPS for heights requires more training for use ofNAVD 88 and GPS and the desirability of enhanced partnering betweenNGS, DOD, DOT, and the private sector in GPS use.

Figure 1 Federal Geographic Data Committee (FGDC) Structure

The FGDC is chaired by USGS

-- USER FORUMS


The GPS Interagency Advisory Council (GIAC), chaired by NGS,represents non-navigation aspects of GPS for all civil federalrequirements.

Federal GeodeticControl Subcommittee

(FGCS)

GPS InteragencyAdvisory Council

WorkGroups

Fixed ReferenceStations

Geodetic Controland SurveyingRequirements

Instruments

Methodology

Vertical ReferenceSystems

Figure 2 The Federal Geodetic Control Subcommittee (FGCS)Structure. The FGCS is chaired by NGS


� ��86(5�1(('�$66(660(176

3.1 Application Categories

Nineteen user applicationcategories defined the major,but not total users of elevationdata. These 19 applicationswere grouped in five summarycategories as follows:

1. Public Safety

2. TransportationManagement

3. Infrastructure Management

4. Construction and Mining

5. Agriculture and Natural Resources

In each of these five summary categories, Tables 1 through 5 summarizethe individual application category elevation requirements in three forms:(1) Digital Elevation Models (DEMs), (2) static elevations of fixedfeatures, and (3) real-time kinematic (RTK) elevations of moving objects.In one application (police, E-911, and fleet vehicle tracking), the usersrequire horizontal real-time positioning only; but this report will show thatthese 2-D positioning requirements would also be satisfied as a result ofGPS solutions for other applications that require 3-D RTK positioning.

3.2 Public Safety

“Public Safety” is herein defined to include police, E-911, and fleetvehicle positioning; disaster preparedness and response; floodmitigation; coastal stewardship; and seismic monitoring. Elevationrequirements for these applications are summarized in Table 1.

What do these five functions allhave in common?

They all need improved DigitalElevation Models (DEM) andimproved 3-D positioning offixed and moving objects; andimplementation of NAVD 88 iscritical!

-- USER NEED ASSESSMENTS


Elevation Requirementsby Application:

DEM VerticalAccuracy

StaticElevations

RTKElevations

Police, E-911, Fleet Vehicles N/A N/A Horiz. only

Disaster Prep. & Response 1 m 5 cm 15 cm

Flood Mitigation 15 cm 5 cm 15 cm

Coastal Stewardship 15 cm 5 cm 15 cm

Seismic Monitoring 1 m 1 cm 1 cmTable 1. Elevation Requirements for Public Safety Applications

Digital Elevation Models (DEMs), with vertical accuracy of 15 cm (6inches), are vital to the Federal Emergency Management Agency(FEMA) for proactive floodplain management and coastal monitoringand protection. Real-time kinematic (RTK) GPS elevation surveys arealso critical for these FEMA requirements. Continuous GPS monitoring,with alarms that are sounded when movement exceeds specifiedthresholds, satisfies the 1-cm RTK requirements.

3.2.1 Police, E-911 and Fleet Vehicle Services

Although requirements are primarily for horizontal positioning,differential GPS (DGPS) is vital for police, fire departments,ambulances, buses, taxis, delivery trucks, and other fleet vehicleapplications. Nationwide DGPS (NDGPS), described in Section 8.8 ofthis report, is superior because real-time differential corrections wouldbe available nationwide without individual communities needing toestablish and operate their own local GPS reference station andtransmitter. NDGPS, used with Automated Vehicle Location (AVL) andComputer Aided Dispatch (CAD), enable vehicle dispatchers to quicklyidentify available assets closest to an emergency.

With DGPS (and more efficiently with NDGPS), E-911 dispatchersknow where emergency vehicles are currently located, and the fastestway to respond to an emergency. When a police officer, for example,needs rapid reinforcements, the push of a "send help fast" button on thedashboard would accurately inform the dispatcher where help is needed,without delays for radio communication and verbal directions to thescene. This can be done today without DGPS or NDGPS, but GPSautonomous positioning has errors of 100 meters; such errors couldcause police reinforcements to be sent to the wrong block. WithNDGPS, the AVL would identify the correct street and the accuratelocation on that street.



The U.S. Department of Transportation (USDOT)4 estimated benefits ofover $8 billion, over the projected 15-year life of NDGPS, for publicsafety alone. The National Transportation Safety Board (NTSB)considers NDGPS to be an essential technology for saving lives.

3.2.2 Disaster Preparedness and Response

The Federal Emergency Management Agency (FEMA) has embarked ona full-scale effort to help build safer communities. FEMA’s goalsinclude increasing public awareness of hazards and loss reduction(mitigation) measures, reducing the risk of loss of life and property, andprotecting our nation’s communities and the economy from all types ofnatural and technological hazards.

Three FEMA reports are referenced herein. The first FEMA reportindicates that the overall costs of disasters to the United States hasgrown significantly over the last decade; the average annual losses haveincreased to $13 billion. The good news, however, is that mitigationworks, and many things can be done to reduce the impact of futuredisasters. For example, one FEMA report indicates: “During HurricaneOpal (Florida, 1995), none of the 576 major habitable structures locatedseaward of the Coastal Construction Control Line (CCCL) and permittedby the State under current standards sustained substantial damage. Bycontrast, 768 of the 1,366 pre-existing major habitable structuresseaward of the CCCL sustained substantial damage.”5

A second FEMA report6 identifies and assesses risks for various types ofnatural and technological hazards. Many of those risks, especially flood,coastal erosion, and hurricane tidal surges, are elevation based. Thus,improved elevation data leads to improved risk mitigation.

4 U.S. Department of Transportation, March 24, 1998, Nationwide Differential GPSReport; Washington, DC, p. 64

5 Federal Emergency Management Agency, March 1997, Report on Costs and Benefitsof Natural Hazard Mitigation, Washington, DC, pp. 1 and 33.

6 Federal Emergency Management Agency, 1997, MULTI HAZARD, Identificationand Risk Assessment, A Cornerstone of the National Mitigation Strategy,Washington, DC, p. i.



For floods and hurricanes, 3-D DGPS techniques are routinely used tosurvey the heights of buildings, high water marks, and tidal surge limits.FEMA often produces GPS Elevation Certificates of buildings damagedby floods and tidal surges because such damages are not covered byconventional homeowners’ insurance policies. By eliminatingrequirements for local DGPS base stations for pre- and post-disastersurveys, an NDGPS capability could reduce the cost of such damagesurveys by approximately one-third, saving FEMA several hundredthousand dollars annually in survey costs.

Disaster preparedness and response also needs horizontal data that wouldbe provided most cost-effectively by NDGPS. NDGPS is vital toFederal, state, and local emergency response personnel, becauseearthquakes, hurricanes, and tornadoes often destroy street signs, andstreet addresses are very difficult to determine. Hurricane Andrew, forexample, destroyed every street sign for miles in all directions, making itextremely difficult for relief workers to identify locations. In rural areas,houses may have no street addresses at all, but rely on post office boxesof rural route systems for mail delivery. Without street names oraddresses, GPS coordinates become the addressing scheme of choice,especially for FEMA and Disaster Field Offices, which utilizeGeographic Information System (GIS) technology.

3.2.3 Flood Mitigation

Relevant Facts

According to the same (second) FEMA study7, “over 9,000,000households and $390 billion in property are at risk from flooding” andproperty damage, excluding agricultural losses, has escalated to roughly$2.15 billion per year, mostly uninsured. As of January 1998, there wereonly 3.9 million flood insurance policies in effect. Accurate elevation

According to a third FEMA report8, only one-third to one-half of U.S.floodplains are studied by detailed methods which compute base flood

7 Ibid., p. 136

8 Federal Emergency Management Agency, November 1997, Modernizing FEMA’sFlood Hazard Mapping Program, A Progress Report, Washington, DC, pp. 11 and19



elevations (BFEs). Instead;over half use approximatemethods in which BFEs arenot computed. Furthermore,over 2,700 floodpronecommunities are unstudied.

With high-accuracy DEMsand GPS elevation surveys,explained below, FEMA can:

(1) Establish combinedhorizontal/verticalcriteria for high-accuracy flood risk determinations

(2) Automate the hydrologic and hydraulic (H&H) analyses neededto rapidly and cost-effectively produceaccurate and completeflood hazardinformation for theentire nation

(3) Cost effectively supportimplementation ofFEMA’s Flood HazardMapping ModernizationPlan

(4) Yield the full benefitsof proactive floodplainmanagement.

Discussion

As stated above, the NFIP has not had the support needed to performdetailed flood studies for many of the flood-prone communities in theU.S. and lacks accurate elevation data. Furthermore, many communitieschoose not to participate in the NFIP because they do not want to adoptfloodplain management measures required by the NFIP. In suchcommunities, owners of flood-prone buildings cannot purchase floodinsurance.

Figure 3 Increased Risk Due to LandDevelopment in Flood Plains

Because of the unavailability ofaccurate and affordable elevationdata, the National FloodInsurance Program (NFIP)currently relies on horizontalcriteria for flood riskdeterminations; and the NFIP hasbeen unable to perform detailedflood studies for over half of thefloodplains in the U.S.



With high-resolution, high-accuracy DEMs produced by the new LIDAR(LIght Detection And Ranging) and/or IFSAR (Interferometric SyntheticAperture Radar) technologies, discussed in the Appendix of this report,legitimate flood risks could be accurately determined. Furthermore,automated hydrologic and hydraulic (H&H) analyses could be used torapidly and cost-effectively produce accurate and complete flood hazardinformation for the nation. Consequently, flood insurance studies couldbe rapidly and efficiently updated as conditions change.

Figure 4. Two Houses in St. Charles County, Missouri During the Flood of’93. Photos provided as a courtesy by the St. Louis Post-Dispatch.

With flood insurance, an elevation difference of one foot can meanthe difference between $1,160 and $450 in annual premiums for$100,000 coverage.9

By performing high-accuracy GPS surveys of all buildings in or nearSpecial Flood Hazard Areas (SFHAs), lowest floor elevations arecompared with base flood elevations (BFEs) for that location, fromFEMA’s Flood Insurance Rate Maps (FIRMs), to determine theestimated depth of interior flooding from the 100-year (1% annualchance) flood. Combined with the pre-flood replacement value of the

9 Federal Insurance Administration, Flood Insurance Manual, 1994 Edition, RevisedOctober 1, 1997, p. “Rate 4.”

What a difference one-foot can make



building and the area of the building’s "footprint," flood damage modelscan accurately predict future flood damages, not just for one building,but for every building in or near a community’s SFHA.

Proactive Floodplain Management

Knowing the inevitable damages that will result from future flooding,communities with comprehensive elevation data can proactively:

(1) Convince owners of their true need for flood insurance, based onlegitimate (elevation-based) flood risk determinations

(2) Identify candidate buildings for retrofitting/floodproofing prior toactual floods, rather than wait for post-flood mitigation steps toforce this action

(3) Use the predicted flood damages to determine where it is cost-justified to initiate drainage improvement projects that will lowerthe BFEs for an area

(4) Determine depths of interior flooding and estimate flood damages– rapidly -- when floods actually occur, without needing time-consuming post-flood surveys of individual buildings.

With pre-surveyed elevation data for buildings, flooded communitieswould merely survey the high water marks at a few key locations intown, model the flood water elevations, and compute the actual depth ofinterior flooding for each building. With actual depths of interiorflooding and previously known replacement value and building“footprint” area, it takes very little time to accurately estimate the flooddamages to every building in the community in order to expedite fundingassistance to those who qualify.

Conclusions

The current procedure (SFHAs) used by lending institutions forhorizontal flood risk determinations, which causes 15,000,000 buyersannually10 to pay $25 or more for "in/out” map determination, hasinherent limitations that cannot be corrected without the addition of

10 Federal Emergency Management Agency, November 1997, Modernizing FEMA’sFlood Hazard Mapping Program, A Progress Report, Washington, DC, p. v.



vertical criteria for flood risk determinations. Eliminating the need foreach home-buyer to pay for horizontal "map determinations" of floodrisk would save these taxpayers (albeit not the Federal government) $375million annually, or $5.6 billion over a 15-year period used to documentthe life-cycle benefits of NDGPS and other initiatives discusses in thisreport. Furthermore, the results of using vertical criteria for flood riskdeterminations would be vastly superior.

3.2.4 Coastal Stewardship

According to a FEMA study11 that relied on study data from the U.S.Army Corps of Engineers, approximately 20,500 miles (33,000 km) ofthe 84,240 miles (132,350 km) of U.S. shoreline experience “significant”erosion, while 2,700 miles (4,350 km) are subject to “critical” erosion.There are 260 coastal counties in the U.S. FEMA inventoried 26 ofthese counties (10%) in 1997, including GPS elevation surveys of 45,000buildings within coastal high hazard zones (V-zones, subject to velocitywave action). Thousands of buildings are vulnerable to coastal erosionand/or velocity wave action from hurricane tidal surges. If other coastalcounties are similar, then 450,000 buildings nationwide could be in V-zones. Projected erosion rates along the Gulf and Atlantic coasts aretypically two feet per year, and one foot per year on the Pacific Coast.FEMA needs elevation surveys of all buildings in V-zones for multi-hazard mitigation

In addition to coastal erosion, coastal communities are also impacted bya projected rise in sea level as a result of global warming. According tothe Environmental Protection Agency (EPA)12, this rise could be as highas 1-meter during the next century. When the White House asked EPAfor statistics on the numbers of buildings to be impacted by the predictedsea level rise, EPA first attempted to use USGS DEMs with 30-meterpoint spacing and 7-meter root mean square errors. Next, EPA usedUSGS 7.5-minute quadrangle maps with 10-, 20-, and 40-foot contourintervals. Both methods were unacceptable, and nothing better was

11 Federal Emergency Management Agency, 1997 MULTI HAZARD, Identificationand Risk Assessment, A Cornerstone of the National Mitigation Strategy,Washington, DC, p. 160.

12 U.S. Environmental Protection Agency, 1997 Global Warming, The Probability ofSea Level Rise, EPA Report No. 230-R-95-008, Washington, DC.



available. The White House was informed that no good estimates couldbe provided in answer to the question.

Florida is considering the annual or seasonal funding of LIDAR-generated DEMs of all Florida coastlines in order to monitor coastalerosion that endangers life and property; evaluate changes to beaches,sand dunes, and barrier islands that protect coastal resources; andimprove the accuracy of erosion rates and projected erosion hazard areaswithin which new development needs to be controlled. The footprintsand heights of beachfront buildings would be documented, providinginformation that has a variety of uses, including planning for newdevelopments13 The entire United States needs such DEMs, not just forcoastal protection but also for the variety of other applications discussedin this report.

3.2.5 Seismic Monitoring

GPS equipment and techniques provide a unique opportunity for earthscientists to study regional and local tectonic plate motions and conductnatural hazards monitoring. Modern low-cost, lightweight systems,which can be used in all weather conditions, provide geodetic precision(0.5-1.5 cm) with post-processed solutions on baselines that are tens tohundreds of kilometers long. Additionally, many researchers have usedGPS to monitor volcanic deformation. USGS has been utilizing thesetechniques since 1990 to conduct a number of studies, including themonitoring of a network of points in and around Long Valley caldera.This is a site of volcanic and tectonic unrest that includes a high level ofseismic strain release, rapid ground deformation, and an unusually highflux of magmatic carbon dioxide. The results of the studies demonstratethe usefulness of GPS as a monitoring tool. Such measurementcampaigns can be used to develop models to improve our understandingof volcano-tectonic systems and the hazards they pose.

Relevant Facts

• USGS noted a substantial need for higher precision, higher density ofobservations, greater frequency of measurements.

13 Shrestha, Ramesh L. and Carter, Bill, March 1998, “Instant Evaluation of BeachStorm Damage Using Airborne Laser Terrain Mapping,” EOM (Earth ObservationMagazine), pp. 42-44.



• USGS currently uses vertical data to infer the location, orientation, andslip on active faults, the expansion and collapse of dikes, the locationand volume of magma chambers. These studies are documented innumerous journal publications.

• NASA is currently funding a study on the crustal extension of the U.S.Basin and Range Province using survey-mode GPS geodesy.

• GPS surveys are generally used on all seismic studies. Additionally,GPS is often used for monitoring the movement of active faults, thereason why we need a reliable vertical system based on NAVD 88.Often, it is possible to determine the distances between stations, evenover distances up to several hundred miles, to better than 5 millimeters(about a 1/4 of an inch).

• Because of subsidence and liquefaction, port facilities pose specialproblems for monitoring seismic activities, for example, finding astable benchmark within the Port of Long Beach is a problem forseismic monitoring.

• Presently, the Port of Long Beach is establishing an automatedmonitoring system to study the impact of earthquakes on wharfstructures. Because of liquefaction of soil, these structures are subjectto damage during a major earthquake. The Port is interested indeveloping a system to monitor movement with structures in real-timeduring earthquakes, and developing better understanding of designmeasures to minimize damage. The instrumentation, installed onwharf structures and pylons, will consist of slope inclinometers,accelerometers, and piezometers; these instruments will be connectedto a central computer. Two of the primary features the system willmonitor are the lateral (horizontal) movement and the differential(vertical) movement of the features. GPS can be integrated with thesystem as one of the monitoring instruments; it could detect movementin three axes.

• Good vertical control is vital in locating and documenting damageafter seismic events.

Discussion

The availability of a precise and reliable vertical reference datum iscritical both for monitoring the on-going movement of the earth in areasof high seismic activity, as well as locating and assessing damage



following major events. The City of Los Angeles, for example, hadmany problems and a great deal of uncertainty in evaluating locations ofsubsurface damage to utilities following the Northridge Earthquake of1994. Major vertical change is a direct indication of subsurface damage.Being able to quickly monitor change is key to emergency response.Many of these problems had their root causes in the fact that there werenot only different vertical datums within the Los Angeles basin, but thata great many of the vertical reference points (bench marks) the engineersand scientists were relying on had themselves become doubtful. Thiscondition, which results from natural or artificially caused groundsubsidence, is worsened when reference elevations are in error becauseof years of neglect.

Additionally, port facilities are areas of high risk during major seismicevents. Due to liquefaction associated with the geology of a port orharbor location, it is absolutely vital to monitor all components of earthmovement, including height. The added safety associated with the datawould provide a higher level of safety and security to world trade.

Conclusion

Most users felt that it would bebest if the NGS continues toconduct, archive, and distributeaccurate and up-to-date heightinformation on a yearly basis.Also, users felt that NGS shouldcarry out research to foster the useof survey-mode GPS, real-timekinematic (RTK) GPS, LIDARand IFSAR. The NGS, NASA,USGS, and NSF are allstakeholders in the need toimprove the geodetic data availableto the scientific and engineeringcommunities.

America will have cities safer from earthquakes if we obtain higher-quality, denser, and more precise geodetic data. This is critical to thedevelopment of design standards for public works structures andbuildings, in order to make them more earthquake resistant and safer.The failure of a parking structure at a veteran’s hospital during the

Figure 5 GPS Receiver in Use forSeismic Monitoring

Solar powered GPS receivers andantennas measure millimeter-to-millimeter movement of the earth’scrust in the Nevada desert



Sylmar earthquake of 1971 accounted for 108 dead. The collapse ofInterstate 880 during the Loma Prieta earthquake of 1988 caused thedeath of many motorists. The toll would have been considerably greaterif a World Series game was not in progress. The cost from the LomaPrieta and Northridge earthquakes to California’s transportation networkalone has exceeded 10 billion dollars. This does not take into accountthe loss of life, injury, and human misery caused from these recentquakes. More and greater earthquakes are merely a matter of time.When and where is critical. Again, precise and reliable spatial data givethe seismologists and engineers the tools to develop reliable predictionalgorithms, and to develop cost-effective earthquake resistant designs.All who live in earthquake and volcano-prone areas of the nation—California, North Carolina, Nevada, Utah, Idaho, Alaska, Missouri,Arkansas, Oregon, and Washington—will receive direct benefit.

3.3 Transportation Management

“Transportation Management” is defined herein to include marinenavigation and safety; air navigation and safety; vehicle positioning andsafety; and train positioning and safety. Elevation requirements for theseapplications are summarized in Table 2.



StaticElevations

RTKElevations

Marine Navigation & Safety 15 cm 2 cm 5 cm

Air Navigation & Safety 1 m 2 cm 15 cm

Vehicle Positioning & Safety 15 cm 2 cm Horiz. only

Train Positioning & Safety 15 cm 2 cm Horiz. onlyTable 2. Elevation Requirements for Transportation Management Applications

High-accuracy DEMs are required for Intelligent Transportation System(ITS) and Positive Train Control (PTC) applications. The landtransportation community needs RTK horizontal positioning, but notRTK vertical positioning. High-accuracy digital bathymetry is requiredfor marine navigation and safety; GPS-occupied tidal stations are neededto link bathymetry to the ellipsoid height system. All transportationapplications require fixed features to be surveyed at the 2-cm accuracylevel. The nautical community needs RTK positioning, at the 5-cmlevel, for positioning of dredges as well as ships, and to keep vesselsfrom running aground. The aviation community needs RTK positioningof aircraft, at the 15-cm level, for final approach and Category 3



landings; whereas the elevations of the runway needs to be surveyedwith 2-cm accuracy.

In addition to being an enabling technology for Automatic VehicleLocation (AVL) and Computer Aided Dispatching (CAD), differentialGPS (DGPS) is also an enabling technology for the IntelligentTransportation System (ITS), as well as railroad, marine, and airnavigation and safety. NDGPS can be used for control of cars, trucks,trains, dredges, ships, and private aircraft, without need for each user toestablish single-purpose GPS reference stations. DOT estimates thatthousands of lives can be saved annually by NDGPS through preventionof highway and railroad accidents alone.

3.3.1 Marine Navigation and Safety

Most of the port facilities inthe United States are affectedby tidal conditions that makenavigation with largecontainer ships difficult andhazardous. The ships need tobe aware of under-keelclearance (relative to thechannel bottom) and over-head clearance of the ship’ssuperstructure (relative tobridges). Since timing ofarrival and departure is criticalin all shipping operations, it isimportant to have reliableheight information on a real-time basis for the height ofwater and bottom ofnavigational channels. Correct height information can limit use ofballast with ships, an operation where water is used to lower the ship todraw more water. This operation is not only time consuming, but is alsovery expensive. Correct height information can also limit dependencyon tidal conditions; waiting for the right conditions can causeunnecessary delay. In addition, height information is needed duringdocking operation because adequate clearance is needed for theoperation of the crane for loading or unloading containers.



The U.S. Coast Guard (USCG) and the U.S. Army Corps of Engineers(USACE) have developed their current DGPS stations to support thenavigation and safety needs of the maritime industry. Virtually all thoseinterviewed expressed the attitude that the Federal DGPS stations havebeen a good investment as a basis for NDGPS. The USACE uses DGPSto accurately position dredges and to ensure that dredging is performedat the correct locations and to the required depths. Currently, ships relyon DGPS to determine their horizontal positions, but not theirelevations. These DGPS stations will require a reliable height system toprovide accurate vertical data.

Relevant Facts

• Marine navigation is currently dependent on a number of differentheight datums and definitions. Navigation of vessels, today andtomorrow, requires a common and reliable height definition.

• Use of various height datums is very confusing and can impact marinesafety. Use of new navigational systems is dependant on GPS, radiolinks, and other motion sensors. Differential GPS (DGPS) can providethe pilot real time information on under keel and overhead clearancefor the vessels entering a port with full cargo, or leaving the portempty. Using GPS, combined with other sensors, the pilot is able todetermine the squat factor and the tilt of the vessel, and, thus, avoiddamaging the keel.

• The GPS navigational instrumentation, used by the pilots and shippingindustry, is dependent on reliable height data. Currently, the pilots areusing heights derived via differential correction from the local verticaldatum.

• Because there is currently no national height system, many ports,including the Port of Long Beach, have established a local geoidalmodel to deal with subsidence and hydrographic surveys.

• Port facilities are required to publish height information using theMean Lower Low Water (MLLW) datum. Using this datum, heightvalues are normally positive numbers. This might result in problemswhen different datums are in use in the same geographic area. Forexample, the City of Los Angeles publishes height values using MeanSea Level (MSL) datum, while the Port of Long Beach publishes inMLLW. The difference between these vertical datums is clear to thePort of Long Beach surveyors, but may be confusing to others.



• The GPS is the “core” component of a modern marine navigationsystem providing navigational information on a real time basis. GPSnetworks were established to monitor height differences within someports, a task that has been completed in past years by leveling.Leveling will still be part of the monitoring, but in a reduced role.

Discussion

The entire marine industry, including the Marine Academy, pilots, andshipping owners, must be involved in modernizing the navigationalsystems. The process must include a provision for education of newpilots and continuing education of current pilots and others involved inmarine navigation.

There is currently no estimate on the cost of integrating a new heightsystem with currently available marine navigation tools. Any costs toestablish a modernized height system, and assure a safe marinenavigational system, should be carried by all levels of responsibility,including the Federal government, shipping owners, and mariners.

Working with the Port of Oakland, NGS has proven that DGPScan provide real-time measurements of a vessel’s settlement,squat, trim, roll, pitch, and heading. Furthermore, DGPS canprovide the position of a vessel’s keel in real-time to within 10centimeters (4 inches) relative to the bottom of the shippingchannel. This clearance is critical. Ships that only barely touchthe channel bottom are stopped for several days for mandatoryinspections. Depending on cargo, every additional inch of draftcan be worth tens of thousands of dollars to shippers per voyage,so shippers are tempted to load their ships to the maximum. Portswith shallow channels lose business to competing ports withdeeper channels. Representatives of the maritime industryunofficially estimate that NDGPS reference stations near portsand harbors would cause an annual increase of $16 billion incargo value in domestic waters, and an annual increase of $640million in tax revenue, or a $9.6 billion benefit over the projected15-year life of the NDGPS.



Any action to improve marine navigation will also improve safety andefficiency. Free and safe passage of freight will benefit the local andnational economy. On the other hand, a major accident within themarine industry due to navigational errors has the potential for disastrousconsequences, with loss of life and loss of economic benefits.

Conclusion

The future of safe marine navigation is dependent on a modernized NHS.Local solutions are not adequate to meet the needs to deal with coastaland international navigation.

3.3.2 Air Navigation and Safety

Regardless of the GPS augmentation system used, high-accuracy DGPSsurveys, called Area Navigation Approach (ANA) surveys14, are used for3-D surveys of Primary and Secondary Airport Control Stations (PACSand SACS), which provide control for various forms of airport surveysrelative to NGS’ Continuously Operating Reference Stations (CORS).Since CORS are assumed to have zero errors relative to the earth’scenter, airport surveys relative to the CORS effectively provide absoluteaccuracy rather than relative accuracy. DGPS procedures have beenproven to be feasible for Category 3 (zero visibility) landings, as well asCategory 1 and 2 landings. Furthermore, ground-based pseudolites15

(essentially, GPS satellite transmitters that operate from fixed locationson the ground) could be well suited for such applications, but there areno official programs for pseudolite initiatives. The FAA is developingan alternative Wide Area Augmentation System (WAAS) for airnavigation and traffic control purposes, designed solely for air safety.The WAAS uses a downlink from geostationary satellites with line-of-site communications to aircraft and air traffic controllers. Both the Wideand Local Area Augmentation Systems (WAAS and LAAS) arediscussed in greater detail in the Appendix.

14 Federal Aviation Administration, September 1, 1996, Standards for AeronauticalSurveys and Related Products, FAA No. 405, Washington, DC, pp. 3.1-3.10.

15 Cobb, Stewart, and O’Connor, Michael, March 1998, “Pseudolites: Enhancing GPSwith Ground-based Transmitters,” GPS World, pp. 55-60.



3.3.3 Vehicle Positioning and Safety

The Nationwide Differential GPS (NDGPS), Automated VehicleLocation (AVL), and Computer-Aided Dispatch (CAD) capabilities areideal for fleet control of buses, taxis, delivery trucks, etc. Cars equippedwith GPS receivers are extremely popular in Japan and are becomingpopular in the United States as well. This translates into thousands ofnew jobs for Americans. Current autonomous GPS procedures enablepositions to be determined to 100 meters; with NDGPS, vehicles can bepositioned in real-time with accuracy of several meters, with thevehicle’s current location accurately shown on a digital road map on thedashboard. This capability also helps in dispatching police cars and towtrucks to accident sites. Furthermore, ITS, which relies heavily onDGPS technology, will enable many transportation innovations to beimplemented, promoting time and cost savings, and public safety. Forhighway applications alone, the USDOT estimates NDGPS potentialbenefits of $8.388 billion over a 15-year period. DGPS applications forland navigation and vehicle tracking are discussed in greater detail inthe Appendix.

The Caterpillar Corporation estimates $60 billion per year in saving as aresult of an estimated 12% increase in efficiency allowed by the use ofRTK DGPS controlled construction equipment and survey techniques.Such increased production efficiency could benefit the U.S. $26 billionmore in constructed transportation assets under the new 1998 ISTEAappropriation.

According to the Department of Transportation, auto accidents areexpected to be the number 1 cause of deaths in America by the year2020. Today, they rank fourth. ITS will promote transportationsafety.

3.3.4 Train Positioning and Safety

The railroad industry has a potential to save millions of dollars everyyear once NDGPS is available to support the implementation of PositiveTrain Separation (PTS) and Positive Train Control (PTC). TheChairman of the National Transportation Safety Board, Jim Hall,reported in a public hearing that during the first half of 1996 alone, PTSand PTC could have prevented 35 railroad accidents resulting in 26



fatalities, 438 injuries, and over $60 million in damages. The FederalRailroad Administration (FRA), Project Sponsor for DOT’s currentNDGPS initiative, conservatively estimates NDGPS benefits of $67.34million, over the estimated 15-year life-cycle of NDGPS, to the railroadindustry alone. Combined with benefits to other industries, NDGPShas a benefit-cost ratio of 152:1.16

3.4 Infrastructure Management

As defined herein, “Infrastructure Management” includes water supplyand quality; subsidence monitoring; and stormwater/utilitiesmanagement. Elevation requirements for these applications aresummarized in Table 3.

Elevation Requirementsby Application


StaticElevations

RTKElevations

Water Supply & Quality 15 cm 2 cm 5 cm

Subsidence Monitoring 2 cm 1 cm N/A

Stormwater/Util. Management 15 cm 2 cm 5 cmTable 3. Elevation Requirements for Infrastructure Management Applications

High-accuracy DEMs are vital for management of America’sinfrastructure. The Geographic Information Systems (GIS) for moststate Departments of Transportation need elevation data for upstreamand downstream ends of culverts, that pass under state roads andhighways, for management of drainage systems as conditions change.Every community needs good elevation data as a high priority, because itis vital to the provision of water for local citizens and for stormwaterdrainage. Changes in elevation caused by subsidence are also vital formany communities, because subsidence threatens the continued supplyof water. The 5-cm RTK elevation surveys are required for efficientsurvey of water, sewer, and drainage features.

3.4.1 Water Supply and Quality

Everyone knows that a sustained supply of safe drinking water ismandatory worldwide; but few Americans stop to realize the importance

16 U.S. Department of Transportation, March 24, 1998, Nationwide DGPS Report,Washington, DC, p. 64



of elevation data for wellhead protection, water supply and quality, floodprotection, and management of natural systems that support hydrologicaland ecological functions.

Relevant Facts

The 1986 Amendmentsto the Safe DrinkingWater Act (SDWA)17

established a WellheadProtection Program(WHP) to protectground waters thatsupply wells and wellfields that contributedrinking water to publicwater supply systemsserving 50% of allAmericans and 95% ofrural America. Thewellhead protection area(WHPA) is “the surfaceand subsurface areasurrounding a waterwell, or well field,supplying a publicwater system, through which contaminants are reasonably likely tomove toward and reach such water well or well field.”

To comply with the SDWA, each State’s WHP must:

• Delineate the WHPA for each wellhead

• Identify sources of contaminants within each WHPA

• Develop management approaches to protect the water supplywithin WHPAs from such contaminants

17 U.S. Environmental Protection Agency, April 1989, Wellhead Protection Programs:Tools for Local Governments, EPA/440/6-89-002, Washington, DC, pg. 3.

Currently, states and local communitiesare required to comply with the SafeDrinking Water Act of 1997 (SDWA).However, they lack the elevation datanecessary to effectively execute theirresponsibilities. The Southwest FloridaWater Management District considerselevation data so critical to its mission thatit is paying between $5,000 and $7,000 persquare mile to obtain 1-foot contour datacompiled photogrammetrically. This isroughly comparable to the LIDAR-generated DEMs, discussed later in thisreport, which cost approximately $500 persquare mile when mass-produced.Availability of high-accuracy DEMs to thethousands of water agencies throughoutthe United States will undoubtedlyimprove services and reduce costs to thewater consumer.



• Develop contingency plans for each public water supply systemto respond to well or well field contamination

• Site new wells properly to maximize yield and minimizepotential contamination

Without high-resolution, high-accuracy DEMs, compliance with thismandate is essentially impossible.

Discussion

NAVD 88 values for survey control are not available in many areas.While NAD 83 horizontal positions are easily determined mosteverywhere, the same is not true for vertical.

For example, the California Department of Water Resources is currentlybuilding a small pipeline project in San Bernardino County. The NAD83 horizontal positioning was determined in a cost-effective mannerusing GPS and the California Department of Transportation(CALTRANS) High Precision Geodetic Network (HPGN). The HPGNcontrol points, equivalent to HARN stations, have GPS-derivedhorizontal positions; some have low accuracy vertical values. None ofthe HPGN points in the vicinity of the pipeline project have verticalvalues. Bringing NAVD 88 values into the area was too costly for theproject, and a decision was made to use available NGVD 29 elevationswith NAD 83 horizontal positions. Thus, high accuracy horizontalcoordinates were merged with low accuracy elevations, diluting theoverall effectiveness of the surveys. If there had been more economicalmeans of providing NAVD 88 elevations, they would have been thepreferred alternative.

3.4.2 Subsidence Monitoring



Land subsidence, the loss of surface elevation due to removal ofsubsurface support, occurs in nearly every state in the United States.Subsidence is one of the most diverse forms of ground failure, rangingfrom small or local collapses to broad regional lowering of the earth’ssurface. The major causes of subsidence include: (1) dewatering of peator organic soils, (2) dissolution in limestone aquifers, (3) first-timewetting of moisture deficient low density soils (known as hydro-compaction), (4) the natural compaction of soil, liquefaction, and crustaldeformation, (5) subterranean mining and withdrawal of fluids(petroleum, geothermal, and ground water).

During the recent five yearsof drought, the CaliforniaDepartment of WaterResources estimates thestate’s aquifers were beingoverdrafted at the rate of 10million acre-feet per year.Unfortunately, the results ofoverdrafting aquifers has ledto many problems caused byland subsidence, including:

• Changes in elevation andgradient of streamchannels, drains, andother water transportingfacilities

• Damage to civilengineering structures--weirs, storm drains,sanitary sewers, roads,railroads, canals, levees,and bridges

• Structural damage toprivate and publicbuildings

• Failure of well casingsfrom forces generated bycompaction of fine-grained materials in aquifer systems

Approximate levels of subsidence. The signsshow the position of land surface in 1925,

1955, and 1977. Although the rate ofsubsidence has decreased, the continuedpumping of ground water has resulted in

additional subsidence in the past 20 years.

Figure 6 Subsidence in California’s Central Valley



• In some coastal areas, subsidence has resulted in tidal encroachmentonto lowlands.

Relevant Facts

• In many areas of California and other states, the elevation data areobsolete and incorrect because of earthquakes, other crustal motions,and subsidence caused by withdrawal of groundwater and petroleum.

• NGS completed most major leveling projects in the 1960s through the1980s, and many benchmarks have since been destroyed or subject tosubsidence.

• Today, leveling between existing benchmarks in areas of subsidencereveal significant discrepancies between related benchmarks. Also,historical published vertical data (through the early 1970s) indicatesubstantial subsidence throughout the California Central Valley andother areas.

• The use of leveling to maintain vertical networks is labor intensive andcost prohibitive. With appropriate standards and procedures, GPSsurveys provide a cost-effective method of establishing andmaintaining vertical data. Organizations that have the responsibility formonitoring subsidence need the ability to rapidly and accuratelydetermine orthometric heights over a large area.

• The USGS has abundant data and reports of subsidence caused fromgroundwater withdrawal. However, little or no recent elevationdocumentation is available overlarge areas. This is especiallytrue of the San Joaquin Valley inCalifornia.

• Leveling is too expensive formost state agencies responsiblefor subsidence monitoring. Theavailability of a national heightsystem would make the use ofGPS feasible, both technicallyand economically.

Cost of Subsidence Mitigation

$0

$50

$100

$150

$200

Mill

ion

s

Houston-Galveston

Santa Clara County

Florida

New Orleans

San Joaquin Valley

Figure 7 Annual Costs of SubsidenceMitigation



Discussion

The California Department of Water Resources estimates that it will costapproximately $1 million per year to establish an elevation networkthroughout the San Joaquin Valley for subsidence monitoring purposes.Numerous Federal, state and local agencies would share the cost.Federal agencies would include NGS, USGS, U.S. Bureau ofReclamation (USBR) and the U.S. Army Corps of Engineers (USACE).State agencies would include the California Department ofTransportation and Department of Water Resources. Local agencieswould include water districts, reclamation districts, and counties.

According to estimates by the USGS, not implementing such anetwork has an estimated annual cost of over $180 million forsubsidence mitigation in the San Joaquin Valley alone. SeeFigure 7.

Nationally, actual costs of damage caused by subsidence and thesubsequent costs to mitigate and prevent further damage are difficult tocalculate. Figure 8 provides relative estimates on a state-by-state basisof initial damages caused by various forms of land subsidence. Inaddition to the damage estimates presented graphically, the NationalResearch Council18 conservatively estimated the annual costs due toincreased flooding and structural damage to be in excess of $125million. These estimates do not include loss of property value due tocondemnation, and they do not consider increased farm operating costs(re-grading of land, replacement of pipelines, replacement of damagedwells) in subsiding areas. It is estimated that annual subsidence costsmay be about $400 million nationally. As shown in Figure 7, annualcosts of subsidence mitigation in selected areas are estimated as follows:(1) over $180 million per year for the San Joaquin Valley, California; (2)over $30 million per year for Santa Clara County, California; (3) over$30 million per year for the Houston-Galveston, Texas area; (4) $30

18 National Research Council, 1991, Mitigating Losses from Land Subsidence in theUnited States, Washington, DC, National Academy Press



million per year for New Orleans, Louisiana, and (5) $10 million peryear for the State of Florida.19

Conclusions

GPS, combinedwith a precise,reliable, andaccessiblevertical network(NAVD 88), willprovide a cost-effective methodto accuratelyestablishelevation dataover large areas.Specific areas ofconcern are areasin whichsubsidence due togroundwaterextraction, such

as the SanJoaquin Valley, isan on-going serious problem. The ultimate goal is to be able to rely on anational framework network, that would help state and Federal agenciesbetter manage problem areas at a significantly lower cost to taxpayers.

3.4.3 Stormwater and Utility Management

Of all utility systems, stormwater management has the greatest need foraccurate DEMs for efficient hydrologic and hydraulic (H&H) modelingof watersheds, streams, and channels. Hydrologic models predictvolumetric concentrations of water from peak events (10-, 50-, 100-, and500-year floods), and hydraulic models compute where those waters willgo and how flood waters will back-up behind undersized culverts andbridges. These H&H models predict the extent and depth of flood

19 National Research Council, ibid..

A. Mining B. Sinkholes

C. Underground Fluid Withdrawl D. Natural Compaction

E. Hydrocompaction F. Drainage of Organic Soils

< $1 Million $1 - $10 Million $10 - $100 Million > $100 Million

Costs

Distribution of subsidence costs in the United States estimated by the NationalAcademy of Sciences (1991). From USGS FY95 initiative "Land Subsidence from

Ground Water Pumping"

Figure 8 National Distribution of Subsidence Costs



waters. They are vital for flood mitigation and proactive floodplainmanagement. The LIDAR/IFSAR DEMs proposed herein would savethe high costs of conventional stream cross-section surveys, and theDEMs would enable the accurate and efficient automation of the H&Hmodeling process. The nationwide acquisition of LIDAR and/or IFSAR-generated DEMs would save an estimated $200 million in conventionalsurvey and H&H costs.

Stormwater, sewer, and water utilities all need accurate 3-D locations ofunderground pipes, manholes, drain inlets, fire hydrants, and similarutility features. DGPS surveys of the tops of manholes, grate inlets, etc.,combined with manual measurements of invert offset elevations, arecommonly used for GIS databases used in modern utility serviceagencies. In most cases, the NDGPS would not provide the 5-cm 3-Daccuracy needed for this application; however, DGPS from a localreference station will satisfy this requirement. Alternatively, WashingtonD.C. and many other communities would be pleased to have 1-2 meteraccuracy in horizontal positioning of manholes, fire hydrants, stormdrain inlets, etc. (achievable from NDGPS) and 15-cm (6-inch) verticalaccuracy of such features (achievable from the proposed DEMs).

The City of Miami is currently using RTK GPS technology tobuild a water utility information management system. Miami isflat. Its highest peak is just 40 feet above sea level, and itsgroundwater table is only 3 to 6 feet below the earth’s surface.When it rains, this water table is close enough to get sucked intothe city’s aging sewer pipes, causing pipe and pump stationfailures that spill raw sewage onto city streets and into theMiami River.

EPA issued a directive requiring that the city overhaul itssanitary sewer system by 2002. The use of RTK GPS withaccurate height information will not only help to solve theproblem, but the field-inventory of the City’s water and sewerfacilities is going five times faster and costing up to 50 percentless compared to conventional utility-location surveys.



3.5 Construction and Mining

“Construction and Mining” is defined herein to include infrastructureconstruction (light or precision construction); mining and earth moving(heavy construction); and pipeline construction. Elevation requirementsfor these applications are summarized in Table 4.

The availability of a National Height System makes the use oftechnologies such as Real-Time Kinematic (RTK) DGPS positioningfeasible for a variety of cost-saving applications in the mining andconstruction industries. Construction and mining equipment developers(e.g., Caterpillar Corp.) have proven that vehicles, controlled by DGPSand computer programs and relying on DTM, can achieve 3-D positionalaccuracy of 6 inches. Additionally, this technology offers majorefficiency improvements and reductions in risk to human operators indangerous situations. In some areas, autonomous earth-movingmachines are already in operation.



StaticElevations

RTKElevations

Infrastructure Construction 15 cm 1 cm 5 cm

Mining and Earth Moving 15 cm 1 cm 5 cm

Pipeline Construction 15 cm 1 cm 5 cmTable 4. Elevation Requirements for Construction and Mining Applications

3.5.1 Infrastructure Construction

When 3-D coordinates (northing, easting, and/or height) are provided bythe project engineer, as the locations for construction stakes, DGPS RTKtechniques can be used to place the stakes accurately and efficiently.This process saves considerable time and expense in comparison withconventional construction surveys. Cost savings have not beencomputed by the survey industry, but savings could be in the billions ofdollars annually. This technology essentially exists today, and it couldbe implemented in 3-dimensions, given the existence of a NHS.

Relevant Facts

• State transportation agencies require elevations primarily to design andconstruct state highway facilities. In many areas of California, forexample, the elevation data are obsolete and incorrect as a result ofearthquakes, other crustal motions, and subsidence.



• In most areas of California, the vertical component of the NSRS hasnot been maintained since the early 1970s. Some NAVD 88 levelinghas been accomplished; however, large areas (e.g., the CaliforniaCentral Valley) remain without a reliable vertical datum. Because ofthis, transportation facilities are designed and constructed using localdatums that conform to existing fixed works, rather than to eitherNGVD 29 or NAVD 88. (Note: the reference datum is often calledNGVD 29; however, in actuality, the datum is a local datum becauseof the unreliable NGVD 29 bench marks.)

• The lack of a comprehensive NHS has also resulted in problems withroad and bridge construction in many major cities in the United States.In Los Angeles, for example, of the 120 bridges, there have beenserious engineering problems on approximately 30 bridges because ofdiscrepancies in vertical data. The following three examples providesome background of this problem:

• Franklin Avenue Bridge –The bridge is 6” lower on oneend than the plans indicate,causing $150,000 in claimsfrom the contractor. The totalconstruction cost is $2.5million.

• Sunset Drive Bridge – Thecity had to reestablish verticalcontrol, because informationhad been obtained from a Caltrans benchmark instead of a citybenchmark. The extra cost to the city was $30,000.

• Fourth Street over Figueroa – The grade was off 20 feet from thevertical bench mark used. Piling holes are 20 feet deeper than theyshould be. The extra cost to the city was $600,000 on a $6,000,000project.

In Los Angeles alone, according to a local official, the city estimatesthe total cost for vertical problems on all bridges to be $5 to $7million.

• Different cities have different datums, and cities typically require thattheir datum be used when working in their jurisdiction. Chicago has

High accuracy height information iscritical in bridge construction



approximately a dozen different vertical datums in use for variouspurposes. This creates the need for checking when projects crossjurisdictional boundaries. Rarely, if ever, is this budgeted for in theconstruction bid.

• GPS has been used successfully by surveyors to provide horizontal andvertical project control. Since a majority of highway projects are longand narrow, centered along an existing roadway or a proposed centerline alignment, it is necessary to surround the projects with existingcontrol points where possible. Because of the lack of a reliable heightreference system, all projects require leveling of control points withinthe same time frame as the GPS campaign. Having a primary networksurvey, completed using GPS and leveling, provides a reliablereference network for design surveys providing vertical accuracywithin 2-cm accuracy.

• The Minnesota DOT is presently testing GPS-derived elevationsagainst more accurate elevations derived from precise leveling.Lessons learned include the need for a better geoid model, especially inareas where gravity anomalies are highest. Accurate calibration of theGPS signal phase center on the antennas is also critical.

Discussion

State and local government participants, along with GPS industry andconsultants involved in precise positioning using GPS, expressed acritical need for a dense network of precise GPS Base Stations and 1-cmgeoid modeling data. Generally, the requirements for the standard NHSwere to support other technologies, to improve efficiencies, and to savecosts both to the agencies and taxpayers.

Users commented that state and local agencies, together with the privatesector, need to partner with the NGS to implement the pieces missing forthe height modernization effort. The NGS should assume an active roleas program manager for the establishment and maintenance of amodernized NHS.

The benefits of a modernized NHS would be dispersed broadly.Surveyors, engineers, mapping professionals, utility companies, farmers,home-buyers seeking Federally-guaranteed mortgages, and users ofAmerica’s transportation system—virtually every American—will derivebenefits from NHS implementation.



Ultimately, the taxpayers save, and every American benefits, if theNHS is implemented.

In California, for example, the Department of Water Resources and otherlocal and regional water and power entities will realize significantbenefits from a modernized NHS. Critical to the development ofCalifornia’s transportation improvement projects, a modernized NHSwould provide a consistent inter-relationship between projects. NHSwill also facilitate consistent, statewide Geographic Information Systems(GIS) and exchange of related data.

Conversely, water agencies, port districts, and other entities requiringreliable vertical data will be adversely affected by continued neglect ofthe height system in California. Also, local GISs will not bestandardized or interoperable, hampering the development of regionaland statewide systems and the exchange of GIS data. To some extent, allengaged in engineering, earth science, planning, land development, andgeographic information will continue to be adversely impacted. Costswill increase because of non-standardized procedures and duplication ofefforts. Everyone suffers from less efficiency and a delayedimplementation of height modernization initiatives.

Conclusions

Users consistently stated that the United States is lagging ourcompetitors in realizing the full potential of GPS, a vital systemdeveloped by the United States. Other countries are moving forwardrapidly (Japan, Sweden, etc.) while we are waiting for the leadership andfunding of a national implementation plan for NAVD 88.

Users expressed the attitude that the Federal government should supportnationwide implementation of NAVD 88. State and local agencies, andthe private sector, should participate in densifying and expanding the“framework” system in cooperation with the Federal government.

The implementation of a NHS that would provide a resolution to anumber of height related problems will help to contain costs on futurelarge-scale projects. Less “checking and fixing” will be required to findand resolve discrepancies, as jobs will no longer have a verticaldisconnect. Since there are so many unknown factors, the solutions lie



with the Federal Government and private sector survey firms that canwork to the same standards as the government.

3.5.2 Mining and Earth Moving

Computer-Aided EarthmovingSystems (CAES)20 integrate DGPSwith RTK positioning and control ofconstruction vehicles. Constructionvehicles are equipped with DGPSreceivers, linked to GeographicInformation Systems (GIS) withDTMs updated "on-the-fly" as earth-moving machines make their "cutsand fills." As each vehicle’s onboardcomputer radios GPS-based positioninformation back to a dispatchcomputer, the dispatcher monitors the location and status (full or empty,heading, and velocity) of each vehicle in the fleet, monitors where trucksare waiting to be loaded, and redirects them for maximum efficiency.Using the in-cab display, the earthmoving equipment operator views thedesign grade, with cut or fill requirements. Using a moving cut/fillisopach map as a guide, the operator can minimize push distances andincrease efficiency. After the CAES machine has started its cut, theheading and a long section is displayed on the monitor, providing theoperator with a graphical display of the current topography in relation tothe design surface. As the operator makes the cuts, the onboardprocessor updates the current DTM in real time. This on-the-fly updatingenables the operator to assess excavation progress for each individualpass and maximize productivity by receiving immediate feedback. Allof these features have combined to produce a dramatic increase in theamount of useful work that construction and mining machines canaccomplish each day.

20 Long, James, March 1998, “Black Thunder’s Roar: Mining for Solutions with RTKGPS, GPS World, pp. 23-28.

Modern mining and earth-movingequipment is increasingly guided

by GPS, relying on accuratehorizontal and vertical data



Relevant Facts

• Most mine sites rely on local vertical and horizontal control systemsthat are established by the mining contractor. These systems are localand must be recreated at each mine.

• Since most mining operations are using a local system, all operationsare controlled by it, including GPS.

• In some locations, GPS is currently used for all mine surveyoperations, including RTK GPS systems installed on bulldozers, trucksand graders21.

Discussion

GPS is a vital link in thepresent and future operation ofmines in the United States. Ithelps in all phases of theoperation, from planning torestoration of the naturalhabitat to near its originalcondition. The use of GPS, fortracking locations and forcontrol of mining equipment,requires the accurate heightsystem that the nationalimplementation of NAVD 88would bring.

Real-Time Kinematic (RTK) DGPS positioning has many cost-savingapplications in the mining and construction industries. Construction andmining equipment developers (e.g., Caterpillar Corp.) have proven thatvehicles, controlled by DGPS and computer programs with DTMs, canachieve 3-D positional accuracy of 6 inches, with major efficiencyimprovements and reduced risks to human operators in dangeroussituations. Some autonomous earth-moving machines are already inoperation.

21 Long, ibid

Figure 9 RTK DGPS for Construction andMining



The Caterpillar Corporation estimates $ billions per year in savingsas a result of an estimated 12% increase in efficiency allowed byGPS modernization.

Conclusion

Implementation of the NDGPS, and/or upgrade of future GPS satellitesas recommended below, could eliminate many (but not all) of therequirements for mining and construction companies to operate theirown DGPS reference stations that transmit DGPS corrections to eachvehicle. Instead, DGPS corrections would be received direct from theNDGPS beacons, or not needed at all with effective upgrades to GPSsatellite frequencies. If cost benefits are anywhere near the $billionsclaimed, this should translate into reduced costs for future roadconstruction and other expenses borne by American taxpayers.

3.5.3 Pipeline Construction

As with the other construction sectors, pipeline companies andconstruction contractors are increasingly using GPS as a survey tool forconducting surveys for new pipelines and for locating and maintainingexisting pipes. Accurate location data are especially important forpopulation density survey and for tracking facilities’ locations andmaintenance operations.

Relevant Facts

• Many pipeline companies are not comfortable with the present statusof vertical control. This is especially the case in Southern California.For example, the Southern California Gas Company needs a reliablevertical datum, preferably using NAVD 88 definition, for its servicearea. Federal law mandates that they provide accurate maps for thelocations of their gas lines and services. The inadequacy of reliableand consistent vertical references causes thousands of dollars inadditional mapping costs each year for this utility alone.

• The use of GPS as the tool for engineering and right-of-way surveyshas been increasing in recent years. However, the lack of a reliablevertical datum results in problems with heights. The solutions areusually made on a project-by-project basis, without consideration forthe entire service area. While solving the immediate problem, these



solutions are generally “stop gap” efforts that ultimately result inhigher construction and system maintenance cost that are passed on toconsumers in the form of higher rates.

Discussion

In general, the participants from the gas pipeline industry were satisfiedwith the results of their survey work. However, it would be desirable tohave a more reliable height reference system in support of the operationssince extensive work is being completed to establish a service area wideGIS to manage future operations.

The industry expects NGS to provide specifications and standards on anational basis. GPS is a common tool for most gas and pipelinecompany operations; the surveyors are dependent on good results.Therefore, uniform standards are important for all involved in providingsurveying and engineering services.

3.6 Agriculture and Natural Resources

“Agriculture and Natural Resources” are herein defined to includeprecision farming; forestry; recreation; and environmental protection.Elevation requirements for these applications are summarized in Table 5.

Elevation Requirementsby Application


StaticElevations

RTKElevations

Precision Farming 15 cm 2 cm 5 cm

Forestry 1 m N/A Horiz. Only

Recreation 15 cm 2 cm 5 cm

Environmental Protection 15 cm 2 cm 5 cmTable 5. Elevation Requirements for Agriculture & Natural Resource Applications

Similar to the mining and construction industries, real-time kinematic(RTK) DGPS positioning has become vital for agriculture, forestry,environmental protection, and other applications related to naturalresources. Farm machinery developers (e.g., Case, John Deere) aredeveloping modern farm equipment with GPS guidance and/orautonomous control. All applications require DEMs with accuracybetween 15-cm (6-inches) and 1 meter.



3.6.1 Precision Farming

There is a sea change currentlyunderway in the agriculturalindustry. New informationtechnologies, with GPS data attheir foundation, are changingfarmers’ relationship with theland, bringing them quiteliterally down to earth. The lastrevolution in agriculturaltechnology, the development of chemical pesticides and fertilizers,brought farmers tools for mass-producing crops. This large-scale farmingsystem treated fields uniformly: agricultural chemicals were distributedwithout regard to differences in soil content or plant health.

The benefits of NDGPSfor modern agricultureare well documented22.Precision farmingsystems gather data ontillage, seeds planted,weeds, insect and diseaseinfestations, cultivationand irrigation, and

location-stamp that data with GPS information. Using these data,farmers can micromanage every step of the farming process. Forexample, a farm GIS database might include layers on field topography,soil types, surface drainage, sub-surface drainage, soil testing results,rainfall, irrigation, chemical application rates, and crop yield. Once thisinformation is gathered, farmers can analyze it to understand therelationships between the different elements that affect crop yields.

This new trend in “site specific” farm management is made possible bythe merging of several unrelated technological advances. These includethe personal computer, GPS, GIS, automated machine guidance, infieldand satellite remote sensing, and telecommunications.

22 U.S. Department of Transportation, March 24, 1998, Nationwide DGPS Report,Washington, DC, pp. 60-62.

With GPS-based precisionfarming technology, farmershave been able to go fromfarming by the acre tofarming by the square foot,and they can reduce a majorsource of non-pointpollution.

Perhaps more than any otherapplication, the agriculturecommunity will benefit fromNDGPS. Over the projected 15-year life-cycle of NDGPS,estimated agriculture potentialbenefits total $3.436 billion.



Precision agriculture enables farmers to implement Best ManagementPractices (BMPs) through the careful control of the quantity of water,fertilizer and pesticides placed on different areas of land, dependingupon soil type and condition, slope, and other factors. Height data havespecial relevance because slopes determine the direction in which runoffwill flow, and runoff could adversely impact unintended areas. For thesereasons, the agriculture industry needs good vertical and horizontalcontrol and accurate Digital Elevation Models (DEM).

In recent years, the application of DGPS and DEM technologies has hadthe added benefit of reducing the quantity of herbicides, insecticides, andfertilizer. This has led to a decrease of the adverse environmental impactof these products. Additionally, the more effective use of insecticides,herbicides, and fertilizer has led to the decreased use of these productsand increases in farm productivity, and has a direct impact on waterquality and soil conservation. The reduction in the cost to producefoodstuffs will be carried to the consumer in lower costs for higherquality and safer food.

The fact that spatial data will need better registration may add asignificant cost to the end user. At the present time, agriculture ingeneral is working on a relatively thin margin, and any added costs in thestart-up phases may make the use of precision agriculture prohibitive inmany situations. This has implications for the degree of agriculturalpollution and resource utilization/efficiency. However, in the long-term,the use of accurate vertical and horizontal data, and registered spatialinformation, will providebenefits through increasedproductivity, lower productioncosts, and lower prices for theconsumer.

Conclusions

In the long-term, theprofitability of precisionfarming technology depends onthe development ofmanagement systems that linkinputs applied with yieldsharvested on specific sites.These management systems will be some combination of computerized

Using precise farming technology, fertilizerand herbicide application rates are changedon the go, improving yields and loweringcosts.



decision support systemsand the accumulatedwisdom of experiencedmanagers. History showsthat most of the benefitsof any new agriculturaltechnology go to the earlyadopter. Those who laghave often been forced outof farming. Precisionfarming is expected tofollow the same pattern.Those who begin toaccumulate data andexperience now will be ready to use improved precision technology as itmatures.

Similar to the mining and construction industries, RTK DGPSpositioning has become vital for agriculture, forestry, and environmentalprotection. Farm machinery developers (e.g., Case, John Deere) aredeveloping modern farm equipment with GPS guidance and/orautonomous control.

3.6.2 Forestry

During the estimated 15-year estimated life-cycle of the NDGPS, theU.S. Forest Service (USFS) estimates that the NDGPS will yield netsavings of $6.83 million for command and control of fire-fightingalone.23 With real-time DGPS to support GIS databases and digitalmaps, the fire mitigation plan can be better conceived and executed. Firetanker aircraft, equipped with airborne GPS, could place fire retardantchemicals more accurately and efficiently, avoiding the unnecessaryoverlap of retardant drops over the same areas. DGPS is well suited forthis application since no geographic landmarks are necessary to identifylocation.

Accurate DEMs (NAVD 88) are also needed for forest managementbecause fires burn upward, and computer models of wild fires arebased on slopes, timber height, aspect, and wind direction.

23 USDOT, ibid, pp. 54-56



Using real-time DGPS for controlling retardant drops and targetefficiency will result in 10% savings of retardant mixture. Moreover,DGPS technology enables both ground and air firefighters tocommunicate their locations accurately. This will save lives!

3.6.3 Recreation

Accurate DEMs are required for design and management of golf coursesand ski resorts. DGPS positioning of snow grooming equipment is usedfor grooming to designed snow depths at ski resorts, but NDGPSbeacons would normally be too far away to provide the needed 3-Daccuracy; therefore the most modern ski resorts already operate theirown DGPS reference stations. Casual recreational users of GPSreceivers, with no requirement for elevation data, can use veryinexpensive GPS receivers to enhance their recreational activities. Forinstance, fishermen can use them to determine locations of favoritefishing areas in the ocean, and hikers can use them to locate theirposition in remote wilderness areas. The USACE, National Park Service(NPS) and U.S. Forest Service (USFS) use DGPS surveys, for example,to accurately survey key features in recreational areas and/or to monitorthe location of vehicles that often operate in remote locations. Althoughrecreational users would benefit from high-accuracy DEMs and NDGPS,these benefits are incidental to those obtained for other applicationsindicated.

3.6.4 Environmental Protection

Recognizing the importance of GPS technology, EPA has a GPSWorking Group responsible for developing GPS solutions to EPArequirements. Most requirements from EPA, state, and localenvironmental control officials are for horizontal positioning, but twospecialized requirements for elevation data are identified below.

NDGPS is needed by EPA, state, and local officials to survey horizontallocations of hazardous materials (HAZMAT) incidents, oil spills, andcontaminated water wells accurately and expeditiously. NDGPS wouldexpedite damage assessment, quantification of contaminated areas, andclean-up actions. In the case of HAZMAT spills, time is very critical inassessing the situation and conducting the clean-up process. EPA haslearned that real-time DGPS takes only 50% of the time for doing thesame surveys as when post-processing is used. Also, there is a time



savings associated with returning to and finding a specific site, e.g., forresampling of soils or contamination sites. EPA estimates net 15-yearsavings of $6.33 million, based solely on the benefits of using real-timeDGPS as opposed to post-processing data as at present.

For a second EPA requirement, elevations of water wells are required to1/100th of a foot, exceeding the vertical accuracy afforded by GPStechnology; for this application, leveling will still be required, but it iseffective only over short distances.

3.7 USGS National Mapping Program Evaluations

In 1994, the USGS National Mapping Division conducted an extensiveevaluation of user needs for selected current products. The followinguser evaluations were extracted from the USGS Open-File Report 95-201. See:

http://mapping.usgs.gov/www/external/OF95-201.html

3.7.1 User Evaluations of NAVD 88 – from USGS Open-File Report

Of 2,130 respondents, 41.1 percent of the users indicated that it wasimportant that USGS modify its data to reflect NAVD 88. Organizationswith the highest percentage of requirements for NAVD 88 included: (1)city, town, and local governments, (2) county and regional governments,(3) and private industry. Users with the lowest percentage ofrequirements for NAVD 88 were nonprofit organizations. Users with thehighest need for NAVD 88 indicated GPS applications; users with thelowest need for NAVD 88 indicated sales and marketing applications.Of those indicating requirements for NAVD 88 on USGS products,31.4% indicated that contours and elevation data should be recompiledfrom NGVD 29; and 68.6% indicated that USGS should supply the shiftalgorithms or parameters to allow individual users to shift the data tosatisfy their requirements.



3.7.2 User Evaluations of DEM Accuracy – from USGS Open-File Report

7.5-minute DEM Accuracy Number Percent

1 = Seldom meets needs 27 2.8

2 93 9.6

3 = Sometimes meets needs 343 35.3

4 388 40.0

5 = Always meets needs 120 12.4Table 6. User Satisfaction with 7.5 Minute DEM Accuracy

30-minute DEM Accuracy Number Percent


2 115 23.2


4 111 22.4

5 = Always meets needs 44 8.9Table 7. User Satisfaction with 30 Minute DEM Accuracy

1-degree DEM Accuracy Number Percent


2 124 27.9


4 91 20.5

5 = Always meets needs 38 8.6Table 8. User Satisfaction with 1 Degree DEM Accuracy

These statistics generally substantiate the opinions expressed during theuser needs assessment portions of this NGS study.

3.7.3 Evaluation of DEMs from LIDAR and IFSAR Sensors

In 1994, USGS was asked to perform the Method, Accuracy, Reliability,and Applications Test (MARAT) project to test the Light Detection andRanging (LIDAR) system, under development at the Houston AdvancedResearch Center (HARC) in cooperation with FEMA and NASA, andthe Interferometric Synthetic Aperture Radar for Elevation (IFSARE)system being developed at the Environmental Research Institute ofMichigan (ERIM) in cooperation with the U.S. Army Corps of Engineers



(USACE) Topographic Engineering Center (TEC). USGS was requestedto test the systems because neither FEMA nor TEC considered itappropriate to test their own sponsored systems. This study addressedone of the desires stated by General Gerald E. Galloway of the USACEwho chaired the Interagency Floodplain Management Review Committee(IFMRC) assigned “ . . . to make recommendations . . . on changes incurrent policies, programs, and activities of the Federal Government thatmost effectively would achieve risk reduction, economic efficiency, andenvironmental enhancement in the floodplain and related watersheds”(IFMRC, 1994). General Galloway stated, “As indicated in our report,during the course of our review, we determined that several agencies ofthe Federal Government were working towards development of data thatwould support digital elevation models. These agencies would includeFEMA and NASA, using LIDAR; USGS using conventional mappingmethods and remote sensing; and several elements of the Department ofDefense using overhead platforms.”

The MARAT project covered a 3- x 3-km area centered at the west sideof Glasgow, Missouri, damaged during the 1993 floods. Results are welldocumented24 Since 1994, both LIDAR and IFSAR sensors have madesignificant technological improvements to penetrate tree cover andacquire elevations of the ground rather than tree tops.

24 Canfield, Dan, 1996, Digital Elevation Model Test for LIDAR and IFSARESensors, U.S. Geological Survey, National Mapping Division, Open-File Report 96-401.


� ±�&$6(�678',(6

4.1 Cost Savings from GPS vs. Traditional Surveying/Leveling

This chapter includes a compendium of elevation survey case studiescontaining lessons learned regarding the transition from leveling to GPS3-D surveys. These case studies are summaries of detailed reportssubmitted to NGS from users in the field. Most of these surveys wereperformed prior to NGS’ publication of its GPS elevation surveyguidelines in 199725. Results of these projects often contributed to theformation of or validation of those guidelines, then in draft form at NGS.

25 National Geodetic Survey, November 1997, Guidelines for Establishing GPS-Derived Ellipsoid Heights (Standards: 2 cm and 5 cm), Version 4.3, NOAATechnical Memorandum NOS NGS-58.

Variable Cost Savings from GPS

• Post Hurricane Elevation Surveys 90%

• Post Earthquake Elevation Surveys 66%

• Water District Elevation Surveys 75%

• Crustal Motion Monitoring 99%

• Subsidence Monitoring 45 - 75%

• GPS RTK Construction Surveys 26 - 71%

• County- and City-wide 3-D Control Surveys 26 - 80%

• Topographic Mapping for Reservoir Construction 71%

– CASE STUDIES


4.2 Disaster Response and Recovery

4.2.1 Hurricane Fran (North Carolina, September 1996)

Purpose

To assist FEMA in rapidlyresponding to Hurricane Fran, whichimpacted the North Carolinacoastline on September 6, 1996, four

separate quick-response GPS projects were performed, to 5-cm 3-Daccuracy requirements, including the following:

• Survey high water marks along 150 miles of coastline for modeling ofstorm surge. This was needed to distinguish buildings damaged bywinds (insured by homeowner policies) from those damaged by water(covered by flood insurance only).

• Survey transections across Topsail Island for a revised Flood InsuranceStudy (FIS). An expedited revision was warranted because of majorchanges in topography, including the loss of protective dunes. Therevised FIS would impact construction codes for rebuilding.

• Survey horizontal location, lowest floor and lowest adjacent gradeelevations of approximately 1,000 units, mostly duplexes, on TopsailIsland. These surveys were vital for production of approximately 2,000GPS Elevation Certificates, 24 damage assessment maps, and GISdatabases required by FEMA Region 4 in Atlanta and FEMA’sDisaster Field Office in Raleigh.

• Survey pilings on eleven buildings selected for evaluation of why somepilings failed and others did not.

Accuracy and Cost Comparisons

In four days, GPS static surveys were completed for 54 high water marksalong 150 miles of shoreline, costing $14,530, or $269.07 per point onaverage. At an estimated $1,000 per mile, leveling would have cost$150,000. Therefore, the cost saving from GPS was approximately 90%.

Reported by:Larry N. Scartz, CLSLarry N. Scartz, Ltd.Woodbridge, VirginiaTel: (703) 690-2582

– CASE STUDIES


In seven days, GPS "stand-off" elevation surveys were performed for 457buildings, including lowest floor and lowest adjacent grade elevations foreach. Because of duplexes and other multiple buildings, these 457 surveysresulted in the production of nearly 1,000 GPS Elevation Certificates, forindividual addresses, by Dewberry & Davis, the prime contractor (seeexample at Figure 10). The survey costs averaged $43.15 per building.

In four days, 22 new GPS temporary benchmarks and 135 intermediate“break points” were surveyed for 11 transections which crossed TopsailIsland from the ocean to the intercoastal waterway. The “break points”were surveyed from the temporary benchmarks using a robotictheodolite. The average cost for each of the 157 points was $79.31.

In three days, 139 points were surveyed for the piling study, at an averagecost of $42.82 per point.

Lessons Learned

• GPS is ideal for post-hurricane surveys because local control points areoften buried under sand or rubble, and needed control can be rapidlyextended over considerable distances.

• GPS temporary benchmark pairs, combined with a robotic theodolite,enable a single surveyor to survey hundreds of 3-D coordinates daily.

Figure 10 GPS Elevation Certificate (Post Hurricane)

– CASE STUDIES


Most problems pertained to inaccurate GIS databases and lack ofgeocoding for tax parcels. Even though the GPS elevation surveys werecompleted in one week, it took several months to "sort out" the correctaddress and owner of the surveyed buildings for production of GPSElevation Certificates. This was partly caused by the fact that a new E-911 addressing scheme was being implemented, numerous postedaddresses did not match those in the county’s database, and many addressnumbers were not posted on the buildings (or had been destroyed by thehurricane).

4.2.2 Northridge Earthquake (California, January 1994)

Purpose

To rapidly restore the geodetic controlnetwork necessary for rebuilding roads,bridges, utilities, and other infrastructuredamaged by the earthquake.

Reported by:Fred W. Henstridge, PLSPsomas and AssociatesCosta Mesa, CaliforniaTel: (714) 751-7373

Highlights

One GPS team surveyed 150 miles of tidal surge high water marks infour days. At an estimated one mile per day, leveling would haverequired 150 days, and the client wouldn’t know the horizontalcoordinates (latitude/longitude) of the high water marks. Atapproximately 10% the cost of leveling, GPS provided superior resultsand expedited relief to hurricane victims.

Using GPS “stand off” elevation surveys which combined GPS andtraditional surveying, 3-D surveys for 457 buildings were performed inone week. With traditional survey methods, it would have requirednearly a week to extend control from the two nearest undamagedbenchmarks on the other side of the intercoastal waterway beforetraditional surveys of individual buildings could have begun.

– CASE STUDIES


Background

The January 1994 Northridge earthquake permanently deformed theground surface in the San Fernando, Simi, and Santa Clarita Valleys andin the northern Los Angeles basin. The force of the 6.8 earthquake raisedthe elevation of ground by as much as 20 inches (50 cm) and changedhorizontal positions by as much as 8 inches (20 cm). This caused thenetwork of permanent geodetic and survey control points, used byengineers and surveyors, tobecome distorted and renderedinaccurate. The transportationinfrastructure within the areawas severely damaged andmany of the vital transportationlinks joining major areas werecut and emergency measureswere put in place to reroutetraffic and to plan a long termrepair of damaged bridges,structures, and roadways.After the area was declared a major disaster area, the CaliforniaDepartment of Transportation (Caltrans), with funding from FEMA,started the recovery work, which would last ten months. Topographicand design surveys were immediately carried out to provide engineersthe vital data to plan and design the repair work. During these surveys, itwas confirmed that the essential horizontal and vertical survey controlhad been distorted and made unreliable for these engineering and designpurposes. Therefore, it was necessary to use localized survey control forindividual projects and start planning for solutions to restore the areawide horizontal and vertical geodetic control. Without this precise andreliable control net there would be chaos in all future design andengineering projects.

Approach

To seek a solution to this problem, Caltrans joined in a cooperativeagreement between the National Geodetic Surveys (NGS), U.S.Geological Surveys (USGS), and the Federal Emergency ManagementAgency (FEMA) to rebuild the geodetic infrastructure in the LosAngeles basin. In addition to restoring the geodetic survey control, thisproject provided a unique opportunity to use GPS combined with preciseleveling in a multi-disciplinary scientific study. This would be the first

Highlights.

This project demonstrated thatGPS techniques cost 66% lessthan traditional surveytechniques for reestablishingthe entire horizontal andvertical control networksfollowing the Northridgeearthquake in 1994.

– CASE STUDIES


time such an approach was undertaken in the United States. This projectwould also provide critical data to investigate and analyze thegeophysical aspects of an earthquake using GPS-based geodesy; todevelop standards and specifications for maintaining vertical controlthrough the use of GPS; and to analyze errors in GPS heights relative togeoidal models.

The USGS was the leadagency, responsible forreporting to FEMA,analyzing historicalgeodetic data from 1974,and providing GPSobservations for thehorizontal network. TheNGS was responsible forfirst order levelingobservations, analyzingdistortion in the nationalgeodetic framework anddeveloping standards andspecifications for GPS derived heights and leveling. Caltrans wasresponsible for providing leveling along historical freeway routes,providing GPS observations for vertical control, and providing logisticalsupport to other agencies.

Caltrans’ internal engineering and survey resources were not adequate tomeet the short-term needs of this project. All of the Caltrans staff wasabsorbed in performing damage assessments and conducting emergencyrepairs. To carry out this project in the required time frame, Caltranscontracted with the private sector to provide additional survey staff andtechnical resources. All survey consultants were assigned to variousconstruction sites to support repair and reconstruction activities,including GPS surveys and traditional leveling.

The NGS staff recovered over 1000 benchmarks and performed over1,500 km of leveling during a three-month period. Caltrans installed newbenchmarks on a number of key structures to assist in future engineeringstudies. The leveling, where appropriate, was performed using amotorized leveling technique, developed by the NGS, using first orderspecifications and procedures.

Figure 11 Observations from GPS-basedEarthquake Analysis

– CASE STUDIES


Caltrans recovered and/or established 120 benchmarks along theirtransportation corridors that were suitable for GPS observations. Thisincluded a number of benchmarks used by the NGS for the first orderleveling. All new monuments were installed in bedrock, bridgeabutments and other stable engineering structures. Where this was notpossible, deep-rod monuments were installed in conformance with NGSspecifications. The GPS height observations were conducted using 12 to15 Trimble 4000 SSE dual frequency geodetic receivers with 2.5 to 3.0hour observation times. All baselines were observed twice using adifferent satellite constellation, different survey personnel and receiverseach time. The baselines were reduced using the precise satellite orbitephemeris, and network closures were completed to validate the integrityof the data. Final adjustment of the network was completed with ties tocontinuous tracking GPS stations and Very Long Baseline Interferometry(VLBI) control stations, using the vertical data derived from the NGSvertical control network.

– CASE STUDIES


Findings

• It is critical to have precise and reliable vertical and horizontal surveycontrol on or within major engineering structures and alongtransportation corridors. Proper and timely engineering and designactivities rely on this control.

• statistical data on the performance of the structure immediately after amajor disaster. Precise and reliable vertical data provide the engineerscritical, historical data on the performance of the structure. This isessential for future earthquake resistant designs and retrofitting ofexisting structures.

• Integration of GPS and precise leveling have provided a valuablemeans for Caltrans to reliably and economically maintain and restoregeodetic control if /or when another major earthquake takes place inthe Los Angeles basin. It is estimated that the level of effort requiredto reestablish the entire horizontal and vertical network using GPStechniques is approximately 66% less than using conventional surveytechniques. This major reduction of effort and cost is made possibleby the use of GPS survey technology in conformance with NGSGuidelines for Establishing GPS-Derived Ellipsoidal Heights. TheNorthridge project was one of the projects used by NGS to define thesestandards, published in 1997.

• The number of primary control monuments can be reducedconsiderably by using GPS techniques. Leveling requires a stablemonument every 1-km along a level route. In using GPS for verticalcontrol, the spacing of monuments can be increased to a 5-km interval.However, for project control, it is desirable to have additional controlsince the control values can be published for horizontal and verticalcomponents.

• GPS derived heights are accurate within 2 cm (3/4 inch) using theNGS guidelines. This accuracy is adequate for engineering projectsover large areas since the relative accuracy between control points isgreater than 2 cm. Geodetic leveling is still needed to meet the 1 mmaccuracy specifications for First Order vertical control.

Figure 12 Geodetic Damage/Restoration Following the Northridge Earthwuake

– CASE STUDIES


Conclusions

The use of GPS provides a reliable and cost effective technology for theengineers, geologists, geophysical engineers, mappers and surveyors inassisting with emergency recovery activities associated with majorearthquakes or geological events, such as slides and erosions.

Southern California needs a reliable height reference network to help indevelopment and maintenance of major transportation facilities. Thiswill not only be vital to support present engineering work, but is moreimportant to assist in fast recovery during major disasters.

For the use of GPS derived heights to be possible and effective, theremust be one common, reliable, and accessible vertical reference datum.This eliminates the doubt or ambiguities that can arise when determiningthe effects of a geological event; e.g., did the surface of the earth moveor was the vertical datum or control station in error?

This common reference datum should be based upon a national standard,as opposed to a local datum. This allows the GPS height observations,within the area of the geological event, to be referenced to geodeticheight control stations well outside of the affected area. This is criticalto the understanding of crustal movements and planning for futureevents.

4.3 GPS Elevation Surveys for Infrastructure Management

4.3.1 Metropolitan Water District of Southern California

Purpose

The GPS-derived Ellipsoid HeightsProgram was initiated as an"earthquake insurance policy" for theMetropolitan Water District’s verticalcontrol investment and to guarantee

the reliability and recoverability of the extensive vertical system withoutre-measuring the entire network.

Reported by:Michael A. DuffyMetropolitan Water District of Southern CaliforniaGlendora, CaliforniaTel: (909) 392-2539

– CASE STUDIES


Background

Metropolitan supplies water to Southern California through adistribution system consisting of 775 miles of pipeline, 243 miles ofaqueducts, five filtration plants, eight reservoirs, and 15 hydroelectricpower recovery plants, and (with NGS) had previously run over 800linear miles of level lines to NGS first order standards. Upon receipt ofNGS’ 2-cm Guidelines for GPS-Derived Ellipsoid Heights, Metropolitanbegan re-surveying existing benchmarks to 7 kilometer spacingthroughout the project area because major seismic events couldsignificantly delay capital improvement projects and cost Metropolitan agreat deal of money in down time (about one million dollars a day forEastside Reservoir alone). Aqueducts, canals, pipelines and reservoirsrequire accurate vertical information for proper construction andoperations. Systems such as these must be built to exact specifications toguarantee water distribution at the proper hydraulic gradient to controlpressure and flow rates. The framework for Metropolitan’s verticalnetwork are the Continuous Operating Reference Stations (CORS) andthe High Precision Geodetic Network (HPGN), established by Caltransand NGS.

Accuracy and Cost Comparison

Metropolitan compromised on strict compliance with NGS’ 2-cm GPSguidelines by:

(1) Not using fixed-height tripods

(2) Not collecting meteorological data

(3) Not using precise ephemeris on the network baselines.

Even with these compromises, rms values from GPS-derived orthometricheights (using NGS’ Geoid 96 gravity model) did not exceed 2.5-cmwhen compared with traditional first-order leveling.

Thirty benchmarks covering 150 miles of leveling were used forcomparison. The project report indicated that there was "a time savingsof about a 4:1 ratio by using GPS versus leveling. However, it must benoted that in this particular example leveling does provide a moreaccurate final product. The need for this accuracy over large areas isdebatable for most applications other than water delivery systems."

– CASE STUDIES


Lessons Learned

Regarding orthometricheights, "Metropolitanconcludes that GPS-derived ellipsoid heightsand the Geoid 96 modelare very good tools toderive orthometric heightsand can be used for mostsurvey applications.However, they would notbe sufficient for providingvertical control forconstruction of much ofthe water-relatedinfrastructure."

Regarding ellipsoid heights, "Metropolitan is very satisfied with this newvertical measurement tool because it will help us recover from any majorseismic catastrophes in a short period of time. This can save us a greatdeal of money due to construction delays and survey controluncertainties. We also believe that in time, the geoid model will becomemore and more accurate as to become statistically predictable for use inorthometric height determination for nearly all applications of surveyingand engineering."

NGS’ GPS procedures and guidelines prove valuable to Metropolitan forthree reasons:

• "Reliable elevations can be created in the fraction of the time throughGPS vertical surveys, provided spacing of these benchmarks are keptat about 25 kilometers and the geoid model used is fairly accurate.Proximity to major mountain ranges needs to be avoided.”

• "Accurate geoid heights can be created at a particular point if both theelevation above sea level (orthometric height) and the ellipsoidalheight is known accurately. This information can then be used to re-establish a benchmark's position after a seismic event.”

• "Large areas can be measured after an earthquake with GPS quickly toanalyze the extent of the seismic event and intelligent decisions can be

Highlights

Time savings of about 4:1 wasachieved by using GPS versusleveling. However, levelingaccuracy was superior fororthometric height determination.

Pre earthquake GPS surveysprovide “earthquake insurance” forcost-effective reestablishment of 3-Dcontrol following a major seismicevent.

– CASE STUDIES


Reported by:Jim Davis, RLSConcord Engineering & Surveying, Inc.Concord, North CarolinaTel: (704) 786-5404

made as to where leveling should be rerun over certain regions fordesign and construction purposes."

4.3.2 Greensboro (NC) Storm Water Services

Purpose

To survey and inventory allstormwater infrastructure features,using GPS RTK procedures(required vertical accuracy of 5-cmat the 95% confidence level), to

populate the Stormwater Management GIS. A preliminary requirementwas to establish a survey control network, using GPS static surveys, withone pair of monuments per square mile city-wide (required verticalaccuracy of 3.5-cm), "blue-booked" for entry in NGS’National Spatial ReferenceSystem (NSRS). Leveling wasto be used when points were not“GPS-able” or to resolvediscrepancies detected duringthe QA/QC process.

Accuracy and CostComparisons

GPS static surveys wereperformed for a total of 209monuments. The average cost permonument was $369.42,excluding "blue booking" costs.These control surveys had nounusual difficulties.

Based on the control surveys,RTK surveys were subsequentlyperformed for a total of 8,682storm drainage features. Theaverage cost per RTK point was$13.20. Although horizontalcoordinates proved to be accurate,

Figure 13 Measuring Invert Offsets for a GPSRTK Survey of Sewer Pipes

The upstream and downstreamnode elevations for each sewer pipeare checked in the GIS for correctconnectivity and flow direction.

If the downstream node is higherthan the upstream node, or if theslope of the pipe is too great, theGIS tells us that something is wrongwith our elevation data.

– CASE STUDIES


approximately 2% of the elevations were found to be in error by more thanthree feet compared with the DEMs, for reasons then unknown, causing allGPS heights to be suspect. Such errors were too large to be attributed topotential errors in the geoid model. All storm drainage features wereresurveyed using leveling to assist in the error analyses.

Leveling was performed for all storm drainage features for comparison withGPS-derived orthometric heights. The average cost per point was $15.28,but leveling alone would nothave provided the 3-Dcoordinates needed, so a directcost comparison isinappropriate.

Figure 13 shows the final stepof the stormwater survey andinventory process. After thetop center of the catch basinmanhole cover has beensurveyed in 3-dimensionsusing GPS RTK procedures,the manhole cover is lifted, theinvert offset is measured downto the lowest flow level ofeach pipe inside the manhole.This is necessary forcalculation of invert elevationsfor upstream and downstreamnodes of each sewer pipe.

GIS database attributes areentered into the pen computer,pre-programmed with “pick-lists” for over 20 itemsmaintained in the GIS. Also,digital photographs are linkedto the GIS file for anystormwater feature that is non-standard or unusual in anyway. Other stormwater systemGPS surveys include streamand channel cross-sections,

Highlights

At $369.42 per monument, GPSstatic surveys of 209 surveymonuments achieved 3.5-cmvertical accuracy requirements.

At $13.20 per point, GPS RTKsurveys of 8,692 storm drainagefeatures achieved 5-cm verticalaccuracy requirementsapproximately 98% of the time,and failed to achieve thisrequirement 2% of the time.Compliance with subsequently-published NGS 5-cm guidelineswould have required each pointto be surveyed twice, ondifferent days, at approximatelydouble the cost. Funding wasnot available for this.

At $15.28 per point, levelingwas performed to assist in theerror analysis of GPS elevationsurveys, but leveling alonewould not have provided thelatitude and longitude neededfor the storm drainage features.Therefore, a direct costcomparison between GPS andleveling is inappropriate.

– CASE STUDIES


Reported by:Stephen R. Wolfe, RLSMcKim & Creed EngineersSmithfield, North CarolinaTel: (919) 934-7154

and special bridge and culvert surveys for hydraulic analyses in floodplainmodeling.

Lessons Learned

Because of cost constraints, the firm did not observe each point twice, ondifferent days, as now recommended by NGS’ 5-cm guidelines for GPSelevation surveys, subsequently published in November of 1997. Instead,heights were compared with the city’s DEM as a QC check. This check waseffective in identifying systemic errors, but not in preventing them.

Trimble Corp. evaluated all the data and verified that there had been noinstances of manual over-rides.

Concord concluded two major causes as follows:

• Although most of the DGPS base stations were completely wide open,several of them were not located ideally because of surroundingobstructions. Although surveyed accurately by static surveys, theirRTK corrections were less than satisfactory as satellite geometrychanged. For a few critical minutes, all rover observations were wrongbecause the reference station’s RTK corrections were wrong.

• The rovers were always initialized in the most open area that could befound, then moved into survey points with obstructions. Conditionswere such that results were not always reliable. Their solution forpreventing such problems in the future is to double-check everyinitialization, regardless of the conditions.

NGS is aware that guidelines for RTK surveys needs to be publishedto cover situations such as this. As a “rule of thumb,” redundantoperations are necessary for almost every form of GPS surveying.

4.3.3 Highway/Bridge Construction

Purpose

To establish the primary horizontaland vertical control points for allphases of the design andconstruction of proposed bypasses

on Highway U.S. 70 at Goldsboro and Havelock, NC, including

– CASE STUDIES


photogrammetric control, engineering surveys, right-of-way surveys,design surveys, digital terrain models (DTMs), construction layout, etc.


For 71 control points at Goldsboro and 24 control points at Havelock, rapidstatic procedures (6-10 minutes per point), 15o elevation mask, 5 secondobservation rate, and baseline lengths <5 km were used. Second OrderClass 1 levels were run between 19 selected points:

• For Goldsboro, using two base stations and two rovers, the maximumdifference between GPS- and leveling-derived heights was 9 mm. Theaverage cost per GPS point was $318.

• For Havelock, using one base station and two rovers, the maximumdifference between GPS- and leveling-derived heights was 2.4 cm.The average cost per GPS point was $328.

Lessons Learned

The method used in Goldsboro yielded higher accuracy. The method ofoperating two base units and two rover units simultaneously (a total of fourreceivers) proved to be very productive. This method allowed:

(1) Long observation times between base units (known monuments)

(2) Observation of two vectors to each new point from the differentbase units

(3) Observation of vectors between rover points (typically set as"pairs") -- all concurrently.

Thus the elevation differences between monument pairs at Goldsboro weresignificantly better than for Havelock.

Highlights

For two similar projects, the use of two GPS base stations withthe rovers reduced maximum elevation errors from over 2-cmto less than 1-cm, compared with leveling, at no increase inaverage cost per point surveyed.

– CASE STUDIES


Reported by:Gregory A. Helmer, PLSRobert Bein, William Frost & AssociatesIrvine, CaliforniaTel: (714) 855-5734

4.4 Four-Dimensional Elevation Surveys (x,y,z and time)

4.4.1 Crustal Motion Monitoring

Purpose

To test the possibility of developingNAVD 88 orthometric heights forSouthern California Integrated GPSNetwork (SCIGN) stations to support 4-dimensional (x,y,z and time) crustal

motion monitoring in the state. The project had three targeted goals:

• To test the ability to develop GPS-derived heights over long distances

• To distribute coordinate and elevation data with regional coverage forfurther testing on local projects

• To test the capability of commercial GPS software.

Background

The SCIGN includes 45 existingstations, 45 stations inconstruction, and 250 sites funded.The GPS infrastructure inCalifornia also includes 10Continuously Operating ReferenceStation (CORS) sites, 28 existingBARD (Bay Area RegionalDeformation) sites, 20 Dense GPSGeodetic Array (DGGA) sites, and18 Permanent GPS Geodetic Array(PGGA) sites.

Accuracy and Cost Analysis

The project met its goal ofdecimeter heights over a regionalCORS network. Out of the eightcheck points, only one, at 13-cm,exceeded the target. Examination

Highlights

Leveling would have required32,000 man-hours of surveyeffort, compared with 150 hoursfor GPS surveys, a savings of99%. However, a slope of -0.23parts per million (ppm) southand –0.39 ppm east was inferredfrom the data, indicatingunmodeled slope in the geoid,bias in the datum, or undetectederrors with the GPS receiver.

GPS baselines less than 50 kmhad rms elevation errors lessthan 1 cm. Baselines between50 and 100 km had rms errorsof 2.2 cm. Baselines longer than100 km had rms errors of 3.2cm.

– CASE STUDIES


Reported by:Gregory A. Helmer, PLSRobert Bein, William Frost & AssociatesIrvine, CaliforniaTel: (714) 855-5734

of the Los Angeles and Orange County stations, where baseline lengths areless than 50 km, shows a standard deviation of 9 mm.

A slope of -0.23 parts per million (ppm) south and -0.39 ppm east can beinferred from the data, indicating unmodeled slope in the geoid, or a bias inthe GPS datum, or undetected noise in the GPS vectors.

Horizontal components for all GPS vectors under 100 km were repeatable atthe sub-centimeter level, and compared similarly with NGS coordinatescomputed in 1995 and 1996.

Examination of height differences from repeat vectors showed goodagreement (2.2 cm RMS) for lines under 100 km. Noise for lines over 100km increased to 3.2 cm.

Because of the long distances involved, comparable leveling would haverequired an estimated 32,000 staff hours of survey effort, compared with 150hours for the GPS observations, processing and analyses.

Lessons Learned

California’s established GPS infrastructure of high-precision geodetic dataand processing capacity offers a viable approach to the organization andmaintenance of a statewide four-dimensional spatial reference systemfounded upon the CORS network. GPS data from the CORS, SCIGN, andBARD networks have been routinely used for horizontal control. Thisproject demonstrated the feasibility to use commercial software andequipment, together with a CORS array, to maintain a vertical referenceframe also.

4.5 Subsidence Monitoring

4.5.1 Long Beach GPS Subsidence Network

Purpose

To design a high-precision GPS programto replicate the existing leveling programfor monitoring subsidence in the region ofthe Wilmington Oil Field for the Port andCity of Long Beach, California. The

established precise leveling program, with estimated accuracy of 6 mm,

– CASE STUDIES


includes a semi-annual network of First and Second Order levelingwhich has spanned a period of over 30 years. An interim target was setto obtain GPS-derived orthometric heights within one centimeter relativeto the established Mean Lower Low Water (MLLW) datum at the Port.


The GPS standard deviations of network adjustment residuals were 4mmhorizontal and 7mm vertical. The GPS vertical standard deviation, comparedwith leveling heights, was 6 mm.

In one case, an error of 3 cm inthe leveling data was discoveredby the GPS networks. Thisoccurred with one of the longerwater crossings required toreach Oil Island Freeman. This3-cm error was immediatelysuspect from the GPS data and subsequently proven in the next levelingcampaign.

Comparison between the cost of the leveling network and the GPS networkis presented in Table 9. The leveling network includes a high number ofbenchmark stations, which distorts comparison of the cost per point.

Leveling Network GPS Network

Number of Network Points 317 33

Cost per Point $442 $1,030

Data Collection 1800 staff hours 330 staff hours

Data Processing 100 staff hours 85 staff hours

Total Cost $140,000 $34,000Table 9.Cost Comparison Leveling vs GPS

Lessons Learned

This project demonstrated that sub-centimeter vertical motion and one-centimeter orthometric heights could be determined from a high-precisionGPS network covering a limited area, i.e., 50 km2 and 2.8 km averagebaseline length, when the geoid model is accurate and NGS’ 2-cm guidelinesfor ellipsoid heights are followed.

Highlights:

For less than 1/4th the cost ofleveling, GPS surveys yieldedcomparable or higher accuracy.

– CASE STUDIES


Reported by:Alan Frank, PLSJohnson-Frank &Associates, Inc.Anaheim, CaliforniaTel: (714) 777-8877

andRobert C. Hart, Jr., PLSStoddard & AssociatesLos Banos, CaliforniaTel: (209) 826-5155

4.5.2 Outside Canal and the Delta-Mendota Canal Subsidence Network

Purpose

The purpose of this on-going project is todevelop and use the most cost-effectivemeans for monitoring subsidence activityalong approximately 54 miles along theDelta-Mendota Canal, and 52 miles alongthe Outside Canals, in California’s CentralValley, which deliver vital irrigationwater supplies. This area experiencedsubsidence of over 26 feet between 1926and 1970. This subsidence is mostly a

result of ground water extraction and has been controlled since 1970, butwith moderate and general subsidence continuing. Since the averagegradient of the canal is only 0.4 feet per mile, even mild amounts ofsubsidence could be disastrous to this system. In fact, these canals haveat times failed or been rendered useless as a result of subsidence.


Assuming a single base station (monument 8694) as stable and fixed, 1996GPS surveys (to NGS 5-cm guidelines) for 11 other stations, and the 1996precise geodetic levels, were run through existing benchmarks along theentire 106 mile route. The maximum height differences, when compared tothe 1997 GPS survey was 4.9-cm. However, results were confusing becausethe northern area indicated uplift when relative stability was evident. Thesingle vertical base station 8694 was later determined to have subsided by 3-cm, through GPS surveys from CORS stations. This explained the apparentand illogical uplift of some stations relative to station 8694 when it wasassumed fixed.

Leveling costs are essentially distance-based, whereas GPS costs areessentially point-based. If numerous heights along a level line are required,leveling is the more cost effective. However, if heights are required only atwidely dispersed points, GPS may be more cost-effective.

• For a fee of $26,900, the 1997 GPS observations of 82 stations cost$328 per station.

– CASE STUDIES


• For a fee of $49,200, Second Order Class Two leveling was performedfor 201 kilometers. The leveling cost $245 per kilometer, and the pricewould change very little, whether there were hundreds of stationsenroute, or one at either end of the line. Therefore, if 82 stationswould suffice for the project, GPS would be the more cost-effectivemethod since the average cost per station would be $600 for leveling.

• At $328 per station with GPS, and $600 per station for leveling, GPSyields savings of approximately 45%.

Lessons Learned

When performing subsidenceor crustal motion monitoringsurveys, it is essential to havea stable, stationary mark onwhich to base the survey.Without a stable mark, theresulting values areinaccurate and confusing,and there is no way to tellwhat is really happening.This project re-iterates theneed for more than one stablemark. While this is not anabsolute requirement,without a second mark, there is no check on the first.

CORS sites are a viable alternative to local stable marks. CORS sites, infact, appear to be ideal reference marks as they are clearly outside the localdeformation area. To produce the same type of absolute deformation resultswith leveling would be prohibitively expensive. If a stable mark could befound within 100 kilometers of the project site, leveling to the mark couldcost in excess of $24,500 based on the previously determined cost perkilometer. This cost per kilometer is also based on leveling across flatterrain, while the nearest absolute stable mark would likely be in the bedrockof nearby coastal mountains. As leveling is much more expensive in hillyterrain, this number of $24,500 could very easily be doubled by the time theleveling is done. Meanwhile, using GPS methods at the 5-cm level,references can be made to CORS sites simply by processing the project basestation data, which is a necessary portion of the project, with the CORS site

Figure 14 Subsidence Monitoring Along anIrrigation Channel

– CASE STUDIES



data. Therefore, with no additional fieldwork, ties to stable CORS sitemarks have been made with relatively little additional cost.

Conclusion

GPS vertical surveys are a viablealternative to leveling for subsidencemonitoring under the rightcircumstances. For mostapplications, GPS is considerablyless accurate than leveling; however,many projects do not require thehigh accuracy of precise geodeticleveling. In addition, GPS allowsone to perform subsidencemonitoring in areas where there areno stable marks without performingextensive leveling by allowing one toreference the survey to CORS sites.CORS sites could also be used tobring control to the perimeter of aleveling based subsidence-monitoring project, therebyeliminating the long, expensive runto control, but preserving the relative accuracy on site.

4.5.3 Unstable Dike Monitoring

Purpose

To monitor (monthly) the 3-Dcoordinates of 24 points on an unstabledike, during construction to increase theheight, width, and stability of the berm

on Eagle Island, Brunswick County, NC.


• Because surveys are performed monthly, McKim & Creed have hadthe opportunity to compare the costs for surveying the same points bydifferent methods:

Highlights;

For subsidence monitoring,leveling indicated apparentand illogical uplift in pointsrelative to a benchmarkerroneously assumed to bestable. GPS 3-D surveys fromdistant CORS determined the“stable” benchmark hadsubsided by 3-cm, causingfalse conclusions as to relativeuplift and subsidence.

At $328 per station for GPS,savings were approximately45% when compared with$600 per station for leveling.

– CASE STUDIES


Elevation Survey Method Cost for 24Points

Cost perPoint

DGPS Static Surveys $6,000 $250

Leveling $4,500 $187.50

Combined GPS and Leveling $2,500 $104

DGPS RTK Surveys w/2 BaseStations

$1,500 $62.50

• Compared with leveling, allmethods delivered verticalaccuracy of 2-cm or better.

Lessons Learned

If points are very close together,then static GPS observations costmore than leveling. However,GPS RTK observations producedaccurate results with savings of66% compared with leveling.

If performing a large job withmany points, or if the points arefar apart or hard to get to, McKim& Creed learned that it would bemore cost effective to use twobase stations simultaneously,transmitting RTK corrections ondifferent frequencies fromdifferent known points. The Leica 9500 series GPS receivers have afunction in the controller that lets the rover unit toggle between base stationfrequencies. With this feature, the rover can collect redundant sets ofcoordinates on a point without having to make a return trip to that point asrequired by NGS guidelines.

RTK procedures were adequate to determine that points "C" and "D" atstation 102+00 moved south approximately 0.3 ft. between November 1997and January 1998, but remained stable between January and February.

Highlights:

At $62.50 per point, GPS savingsare 66% compared with $187.50per point for leveling.

GPS is excellent for determiningelevation changes over time.Furthermore, when the GPSrover can toggle between DGPSbase stations, the use of two GPSbase stations, simultaneouslytransmitting RTK corrections ondifferent frequencies, can be acost effective means for obtainingredundant observations. Caremust still be taken to avoidmultipath errors at the rover.

– CASE STUDIES


Reported by: Frank Fitzpatrick Manitou Engineering Co. Escondido, California Tel: (760) 741-9921

4.5.4 Geodetic Control Surveys (1992 and 1998), Escondido, California

Purpose

To provide control for aerialphotogrammetric mapping of a 34 squaremile area, and to establish 326 controlmonuments on a one-half mile grid

throughout the city. The ultimate goal is to build a spatial inventory anddescriptive record of each sewer line, manhole, water line, and gatevalve itemized in a GIS database; this would allow rapid location in anemergency, and routine maintenance could be monitored, scheduled, andreported by querying the computer.

Accuracy Requirements

The city specified First Order Horizontal Control to be spaced at a three-kilometer interval, Second Order Horizontal Control to be a one-kilometerinterval, and Third Order Vertical Control on all monuments

Procedures

In 1992, Manitou performed the survey measurements with single frequencyGPS receivers. The First Order Primary Network was constrainedhorizontally to five State of California High-Precision Geodetic Network(CA-HPGN) control stations and vertically to 14 local benchmarks (2nd and3rd order). The Second Order, Class 1, GPS Network was horizontally tiedand constrained to the entire thirty-one (31) stations in the First OrderPrimary Network, and they were vertically tied and constrained to thefourteen (14) local bench marks cited, plus an additional 38 local benchmarks. Level circuits and reciprocal trigonometric levels were run throughselected GPS stations from the existing benchmarks.

In 1998, Manitou re-observed key tie points with dual frequency GPSreceivers.

Results

The final 1992 base line results indicated that 99.6% of the 23,511 base lineswere better than 20 ppm precision ratio for Second Order, Class 1(1:50,000), and 98% were better than 10 ppm precision ratio for First Order(1:100,000). Of the 326 stations, the error ellipse ranges were as follows:

– CASE STUDIES


• Horizontal: 279 stations between 0.00 and 0.10 feet; 47 between 0.10and 0.20 feet.

• Vertical: 294 stations between 0.00 and 0.10 feet; 30 between 0.10 and0.20 feet; and 2 (HPGN’s) between 0.20 and 0.30 feet.

Costs

• The preliminary layout of the one-kilometer control network and clientcoordination cost $5,000 ($15.34 per point).

• The monument recovery, installation of new monuments, and settingaerial targets (including supplies and jackhammer and air drill rental)cost $20,000 ($61.35 per point).

• The GPS planning and observation (including amortization of the GPSreceivers) cost $64,000 ($196.32 per point)

• The data processing, including downloading receivers, processingbaseline vectors, running loop closures, adjusting the first order andsecond order networks, and report preparation, cost $24,000 ($73.62per point.

• The total cost per point, excluding setting new monuments andrecovering existing monuments was $285.28 per point.

1992 Lessons Learned

In preparing the original task list and estimate of time, the session planningtime was grossly underestimated. The preparation of sky chart with the siteobstructions included greatly decreased the number of missed sessions dueto loss of the required number of satellites. However, this increasedplanning, while well worth the time, increased the estimated sessionplanning by fifty percent.

The volume of paper was staggering. With monument recovery card sessionplanning logs, station observation logs and maps of the observation routesfor each receiver for each day, the data filled twelve 1.5-inch thick three-ringbinders. Efforts were made to sort all the sheets that would be reused forsubsequent observations at the same monuments, but more time andemphasis on this task was necessary.

The amount of organization required to keep supplies on hand for themonument installation crew and the design and time allocation for the daily

– CASE STUDIES


observations and move times kept two people, instead of one as planned,occupied full time.

Fixed height antennae poles should be used in the future to eliminate thepossibility of human error.

The single-frequency receivers produced the required accuracies within themapping area where the maximum vector length did not exceed 6 km.However, dual-frequency receivers would allow reduced observation timesand improved vertical precision. The vertical component of the network ties(12 km to 28 km in length) could not be included in the adjustment becauseof the excessive error introduced when they were set to fixed.

1998 Re-Observations

The ties to the CountywideNetwork and the ties to station1010 in the Escondido Projectwere re-observed in March of1998. The re-observations weremade with Trimble dualfrequency GPS receivers inaccordance with NGS standardsfor GPS elevation surveys at the5cm level. Manitou obtainedNAVD 88 elevations on stationsLomax, SDGPS 03, SDGPS 32,and SDGPS 34 from CALTRANS. The horizontal vectors fit withinmillimeters of the record vectors, indicating there had not been anydifferential crustal movement between these stations. The plot ofearthquake faults for the local area indicates a swarm of faults east of theCuyamaca Mountains in the Anza Borrego Desert and additional faults inthe vicinity of San Diego Bay.

The Horizontal Time Dependent Program developed by NGS revealsapproximately 16 cm (0.52’) of secular movement between the 1991.35station coordinates and the 1995 station coordinate for the permanent trackerat station Monument Peak. This difference in position is the distance thePacific Plate has moved in relation to the North American Plate in the yearsbetween 1991 and 1995. The vertical movement is separate and distinctfrom the lateral movement. The vertical movement is predominant in areasof ground water, oil, or natural gas extraction. The Northridge earthquake

Highlights:

Single frequency GPS receivers,used in 1992, satisfied verticalaccuracy requirements only wheremaximum base lines did not exceed6 km. Dual frequency GPSreceivers, used in 1998, allowedlonger baselines (12 to 28 km) tobe included in the networkadjustment

– CASE STUDIES


Reported by:William H. Bond, DirectorOffice of Information TechnologyBaltimore County, MarylandTel: (410) 887-2441

was both a lateral and a vertical movement, and different from the slowsteady creep characteristics of the secular movement.

Geoid 90 was the gravity model in use when the observations wereconducted in 1992. Comparisons with heights obtained from CALTRANSand the Geoid 96 differences show excellent correlation. The difference inthe orthometric height from 1992 to 1998 indicates approximately 0.65-meter bias. The Geoid separation from David Zilkoski’s unpublishedspecial study for San Diego County indicates a reasonably consistent biasfrom the Geoid 96 separations. The correlation of the CALTRANS leveledorthometric heights with the heights obtained from the GPS measuredellipsoid heights and the Geoid 96 separations indicates Geoid 96 fits thisportion of San Diego County quite accurately. The dual-frequency receiversprovided the increased vertical accuracy, compared to the single-frequencyreceivers, which allowed the network ties (12 km to 28 km in length) to beincluded in the vertical constraints for the 1992 network adjustment.

Based on these test results, the Escondido project could be reprocessed toderive NAVD 88 orthometric heights at the 5-cm accuracy level.

4.6 GPS Control Surveys

4.6.1 Baltimore County, Maryland

Purpose

The primary objective of thecounty’s GPS-leveling project wasto support the overall GIS missionof the county directed to

modernize the information infrastructure. The secondary objective wasto investigate the practical feasibility of GPS procedures to supportgeoreferencing of GIS via high-precision airborne GPS-supported aerialtriangulation, and to study cost and time efficiency of the technique. Thetertiary objective was to see how far GPS could be used forestablishment of vertical control bench marks for other surveying anddevelopmental engineering projects in the county.

Background

Baltimore County enterprise GIS has been, from the beginning, a uniqueproject relying on the integration of a number of state-of-the-art geodetic

– CASE STUDIES


and photogrammetric concepts and techniques for the development ofhigh precision geo-data. The establishment of a geodetic controlnetwork, employing the NGS defined specifications for GPS-derivedelevations, is a classical example of this approach. Consequently, thisproject is representative of the first large scale planned GIS developmenteffort at the local government level in the U.S., to successfully integrateNGS specifications and procedures of GPS techniques for verticalcontrol, substituting for conventional leveling.

The GPS 3-D surveys were planned to provide vertical accuracy at a 2cm level. The County worked closely on this project with NGS toensure the National Spatial Data Infrastructure (NSDI) framework. Thispartnership proved invaluable for the experiments with seashore heightmodel use for airborne GPS, and pioneering GPS technique applicationin GIS, which made a three dimensional GPS control network a reality.

Project Phases

The basic approach for geodetic control was to establish a sparsehorizontal GPS framework of control covering the entire county in phase1 of the GIS project. The density of monuments was planned at adistance of about six to seven miles apart. The county’s minimumhomogenous geodetic control network strategy was to have limitedcoverage for economic reasons to be subsequently densified on a needbasis. The network, comprised of 52 GPS points covering the entirecounty area of 675 square miles, included 26 points in phase 1, coveringabout 125 square miles. Additionally, 25 vertical benchmarks wereestablished by leveling in phase 1 to support photogrammetry anddevelopment projects, and for establishing a geodetic referenceframework.

In phase 2, the strategy was based on a combination of leveling and GPSsurveys. A few ground control points were connected by short levelingloops from nearby existing benchmarks. The remaining control,particularly in difficult hilly and riverine terrain, or where existingvertical control was sparse, was planned using GPS techniques as pilotexperience. If the method proved successful, phase 3 work was to beundertaken with the new procedure. This approach proved beneficial, asonly 33 more new GPS monuments were added in phase 2 and 3 formulti-purpose use.

– CASE STUDIES


Phase 1 and Phase 2

In phase 2, four GPS baselines did not meet the guideline’s 2-cmtolerance limits. All of these baselines were greater than 10 km, whichexceeds the maximum length requirement stated in NGS’ guidelines.This indicates that the NGS requirement to limit the length between baselines to 10 km has validity. Several baselines were re-observed, withdifferent satellite geometry, and the results did satisfy the 2-cmtolerance.

Phase 3

The primary objective of phase 3 GPS surveys was to provide 3-Dpositioning of minimum essential ground control to support airborneGPS supported photo triangulation. The secondary objective was toprovide additional minimum essential densification of control at spacingof about 6 km throughout the 290 square mile area.

A few of the GPS derived elevations were outside the 2-cm accuracythreshold, but they tended to be on the periphery of the project area oroutside it. As such, they did not significantly contribute to inducing errorin the adjusted 3-D positions of the other interior stations. The resultssatisfy requirements of 2 cm level accuracy in this project.

Conclusions

Until recently, reliable vertical bench mark values sufficient to supporthigh accuracy photogrammetric projects, GIS data bases, geodeticcontrol, and a variety of surveying and engineering project applicationscould only be obtained by leveling. While providing good results,leveling is extremely time and labor intensive, and causes delays andhigher costs for survey work. Baltimore County’s GPS pilot project inPhase 2 as well as Phase 3 production fully met the expectations ofachieving 2-cm orthometric height accuracy. Even a slightly relaxedaccuracy level of 2-cm to 4-cm should be sufficient to support the highaccuracy GIS photogrammetry specifications of this and many othersimilar projects. The experience of this project validated the NGSguidelines and procedures for obtaining reliable GPS heights, usable formany engineering and surveying applications.

– CASE STUDIES


Cost Comparisons

The County accomplishedsignificant cost savings of about80% compared with theconventional method of levelingin the final Phase 3 of theproject. The time efficiencywas also remarkable, which wasabout 10 times more efficient inPhase 3 as compared to Phase 1 of the project.

• Phase 1. The cost of leveling for Z, and GPS surveys for X and Y, cost$600 per square mile.

• Phase 2. The cost of some leveling for Z, and some GPS surveys forX, Y, and Z cost $358 per square mile

• Phase 3. The cost of 3-D GPS surveys for X, Y, and Z cost $104 persquare mile.

4.6.2 Eastside Reservoir Project (RTK Surveys)

Purpose

Initially, the purpose of the project wasto use GPS to provide control ofphotogrammetric mapping forconstruction planning, earthwork

monitoring, and volumetric computations. Ultimately, the purpose wasto use GPS RTK procedures to provide real-time monitoring andvolumetric computations for this major construction project.

Background

When the Eastside Reservoir is built by the end of 1999, at an estimatedcost of $1.9 billion, it will be one of the largest water reservoirs in theworld. It will be formed by two earth/rock filled dams 4.5 miles apart,plus a third earth/rock filled dam at the low point in the north rim. Itwill have a storage capacity of 269 billion gallons, cover an area of 4,500acres, and have a varying depth of 160 to 260 feet. The West Dam willbe the largest earth/rock dam in the United States, 8,700 feet long and285 feet high. It will have 9,000 acres set aside for wildlife reserves to

Reported by:Fred W. Henstridge, PLSPsomas and AssociatesCosta Mesa, CaliforniaTel: (714) 751-7373

Highlights:

Compared with conventional 3-Dsurvey control, GPS providedcost savings of 80%. Some GPScontrol surveys were completedin 1/10th the time of prior

– CASE STUDIES


protect and enhance the region’s wildlife and plants. In addition, another1,950 acres of recreational and open space areas are being planned forcamping, hiking, picnicking, riding, fishing, and boating in special lakesseparated from the main reservoir.

The Metropolitan Water District of Southern California (MWD) isbuilding this capital improvement project in support of the expectedgrowth in the Southern California area. This reservoir will double theamount of surface storage available for MWD's service areas and ensurea more reliable delivery of water during peak summer months, droughts,and emergencies. This enlarged water capacity will protect both thequality of life as well as the economy from water shortages that haveplagued other parts of the West.

Approach

MWD contracted Psomasand Associates to providesurvey personnel as anextension of their surveystaff for this five-yearproject. At the height ofthe construction period,the reservoir will employover 500 constructionworkers, includingconstruction managers,engineers, surveyors andarchaeologists. Thesurvey staff plays animportant part inmaintaining a tightschedule during the excavation and placement of some 90 million cubicyards of clay, sand, and rock to build the dams.

At the outset of the project, it was decided to divide the 9,000 acres into1,000-foot square grids and calculate the quantities by aerialphotogrammetric mapping. For this purpose, the surveyors set permanentmonuments and placed aerial targets at selected locations. Wherepossible, these monuments and targets were placed at secure areasoutside construction activity. All survey control was established usingGPS techniques by ties to a primary control network established by the

Figure 15 Eastside Reservoir Project

– CASE STUDIES


MWD before construction. At the beginning of the project, aerial photoswere obtained for the entire project area. This procedure would berepeated at regular intervals throughout the project. For calculation andreporting of construction quantities, it was planned that aerial photos forspecific areas would be completed as needed.

From the project beginning, the surveyors used GPS surveys, includingRTK, to carry out field surveys. A semi-permanent, continuousoperating reference station (CORS) was maintained to providedifferential correction for all field observations for the first year ofconstruction. MWD replaced this station with two permanent CORS atthe east and west ends of the project. These CORS were constructed byinstalling pilasters with antennae and power lines for electrical power.These reference stations will remain as an integral part of a damdeformation system to monitor dam performance.

Photogrammetric Mapping of 400 Acres

Function Labor Costs Time

Setting of aerial control targets 4 man days $2,600 2 days

Aerial photography $500 1 day

Stereo plotting and mapping 116 man days $50,000 58 days

Digital map overlay for quantities 1 man day $800 1 day

Total 10 man days $53,900.00 64 daysTable10. Eastside Reservoir Photogrammteric Mapping Costs

MWD’s original plan for using aerial photogrammetric mapping forquantities was soon replaced by using RTK surveys to obtain the neededquantities in real time. This method soon became the preferred methodfor obtaining all on-site quantities. Table 11 shows the 71% savingsachieved by using GPS.

RTK GPS Surveys of 400 Acres

Function Labor Costs Time

RTK-GPS survey of site 19 man days $12,920 1 day

Office transfer of GPS data to PC 0.5 man days $550 1 day

Preparation of digital model 1 man day $1,100 1 day

Digital map overlay for quantities 1 man day $1,100 1 day

Total 4.5 man days $15,670.00 4 daysTable 11. Eastside Reservoir GPS RTK Mapping Costs

– CASE STUDIES


The RTK surveys were carried out by vehicle-mounted GPS andpersonal-mounted GPS and walking. The vehicle-mounted GPS wasmounted in a vehicle, with a known antenna height, andmeasurements were taken while the vehicle was moving. At costsavings of 71%, GPS accuracy exceeded those of thephotogrammetric mapping.

Findings

• Providing quantities by mapping has proven to be a very timeconsuming and costly method of reporting. Prior to the flight,surveyors will inspect all aerial targets for the subject grid block andreplace and survey all targets found missing. After the sight is ready,the aerial contractor obtains the pictures and compiles mapping.Timing of the aerial flight is dependent on the weather conditions andcan be delayed by days, or by weeks. Final quantities are available tothe construction manager team 30 to 60 days after request.

• Using RTK techniques, the surveyors are able to complete a multiplenumber of grid blocks on a daily basis, and the office surveyor canprocess the data immediately for reporting. Using a number of rovingGPS receivers operated by only one surveyor each, large areas can becompleted within a few days. When the surveyors are working nearmoving heavy construction equipment, there are times when a secondpair of eyes is needed to provide protection.

• Use of RTK has provided a means for visual inspection of wide areasby the surveyors. The surveyors have found small culverts, wells,pipes, and other small important features that are not visible whencompiling mapping from aerial photographs. This visual inspectionand collection of data can provide a means to log data for historicalrecords. It is possible that these records and data files collected by thesurveyors can be used for collection of evidence or recovery of facts insome future case.

• Accuracy of RTK surveys have been found to be very reliable. At theoutset of the project, when some RTK positions were obtained withouta local geoidal model, the heights were subject to errors from 2 cm to10 cm in areas. After modeling the local geoid, the heights were foundto be within 1-cm range. This was approximately 2.5 times better thanheight obtained from aerial mapping.

– CASE STUDIES


• The level of effort associated with obtaining quantity volumes bysurvey methods is considerably less then using aerial mapping. This isevident from the cost of labor using prevailing wages for the trades.The MWD has analyzed costs associated with the two methods and hasfound that savings associated with survey methods exceed 200% of thecosts associated with aerial mapping. This is a large saving consideringthese reports are needed on daily basis over a five-year period.

Conclusions

• Real-time kinematic (RTK)GPS surveys have beenproven to be an efficient andcost effective means ofcollecting vertical datawithin a large constructionsite. Under most conditions,the cost savings aresubstantial over the use ofaerial photography for datacollection.

• The use of GPS requires a reliable height reference system within theproject area. Establishment of this reference system can be costly atthe start of the project if there are conflicts in or lack of verticalbenchmarks. For this project, the MWD had to run miles of levelsfrom known vertical control into the site. Using GPS for this initialleveling would have saved considerable time and money.

• The use of RTK GPS surveys becomes very cost-effective whenpermanent CORS are established at the inception of the project and agood geoidal model is in place.

• GPS survey techniques have provided a reliable tool to the surveyorsand construction managers to collect data. With additionaldevelopment and refining, it is possible to collect these data in real-time, with a direct radio link to a central database. This will furtherincrease the efficiency of the construction industry and provide an edgein the competitive world market.

Highlights:

GPS RTK mapping cost savingswere 71% compared withphotogrammetric mapping, andGPS required less than 10% thetime.

On-site DGPS CORS wereinstrumental in project success.

– CASE STUDIES


Reported by:Earl Cross, PLSCross Land Surveying, Inc.San Jose, CaliforniaTel: (408) 274-7994

4.6.3 Geodetic Control Network, City of San Jose, California

Purpose

To establish a city-wide geodetic qualitycontrol network, following NGS 2-cmguidelines, to be used as base control forsubsequent surveys by City of San Jose

personnel and private consultants, and to support the city’s GIS/LIS basemap development.


After the GPS network was constrained horizontally and the level line wasadjusted, differences in orthometric heights were found to be consistentlyless than 1 cm.

The cost to establish horizontal coordinates and GPS derived orthometricheights on 111 monuments (including 8 primary control points), prepare areport, and file a 9 page Record of Survey of the results with the CountyRecorder, was $500.00 per station. The cost to establish NAVD 88orthometric heights only (no horizontal coordinates) by leveling on 68benchmarks was $420.00 per point.

In February 1998, six base lines between 9 GPS/benchmark monumentswere re-observed by both methods, with differences between unadjustedlevel runs (NAVD 88) and a free adjustment of the GPS observations basedon GEOID96 of 0.0003, 0.0095, 0.0017, 0.0019, 0.0070, and 0.0200 meters(2-cm maximum discrepancy). "It took twice the time to run the levels as itdid to observe the base lines by GPS. In these times of tight budgets andshrinking funding, it becomes apparent that a considerable amount of moneycan be saved by utilizing a properly designed GPS network to establishelevations for most engineering projects."

Lessons Learned

With today’s technology, electronic digital levels, computers, softwarepackages, and other labor saving devices, it still takes a minimum of a three-person crew to economically run a level loop that meets FGCS SecondOrder Class 1 Specifications. On this subject project, we were able tocomplete approximately 4.9 miles per day. On the other hand, the samethree people, using dual frequency GPS receivers with field techniques that

– CASE STUDIES


meet accepted guidelines and specifications can obtain elevations on nine (9)bench marks that are about 1 mile apart (8 miles total) in one working day.The elevations derived from these GPS observations easily duplicate thoseelevation results obtained from leveling.

One of the main lessons we have learned from this project is that if enoughbench marks with properlydetermined elevations are used tomodel a GPS network or line ofstations to be used as benchmarks, the output of GPS versusleveling can be almost doubled byusing GPS methods, withoutsacrificing any accuracy in the result.

4.6.4 Geodetic Control Network, Davidson County, North Carolina

Purpose

To establish horizontal and vertical controladequate to perform aerial triangulationcomputations to develop digital orthophotosand a planimetric base map for DavidsonCounty’s GIS; and to provide additional

control stations to be used by local engineers and surveyors formaintaining the county’s GIS.


All basic vertical control was extended from existing NGS, NCGS, or NCDOT benchmarks and referenced to NAVD 88. NGS’ 5-cm Guidelines forGPS Elevation Surveys were followed, and Geoid 96 was used to computeorthometric heights. NGS’ VERTCON was used to convert NGVD 29 toNAVD 88 orthometric heights.

There were a total of 96 primary control marks determined for this project.The average cost per station was $800.00. This included fieldreconnaissance, visibility sketch, placement of monuments, missionplanning, data acquisition, and computations.

Reported by:Webb A. MorganWebb A. Morgan &Associates, P.A.Asheville, N. CarolinaTel: (704) 252-1530

Highlight:

Productivity can be almostdoubled by using GPS in lieu ofleveling.

– CASE STUDIES



Lessons Learned

GPS is a practical and cost-effective alternative to levelingfor large project areas wherevertical accuracy requirements donot exceed the 5-cm verticalpositional tolerance. The finalresult of the Davidson Countysurvey far exceeded thephotogrammetric control accuracy requirements. The accuracy of the geoidmodel furnished by NGS has a major impact on the accuracy of the resultantorthometric heights.

4.6.5 GPS Control Surveys, Brunswick and Bladen Counties, NorthCarolina

Purpose

To establish three pairs of horizontal andvertical control points for Magnolia GreenGolf Community, and to establishhorizontal and vertical control for

replacement of Bryant Neil Pond Bridge #46 on NC 131.

Accuracy and Cost Comparison

• The golf community DGPS static surveys cost $1,500 per 6 points or$250 per point; conventional surveys cost $2,040 per 6 points or $340per point. Thus, GPS status surveys were performed at cost savings of26% compared with leveling. The difference between GPS derivedelevations and leveling varied between 2 mm and 1.3 cm.

• The bridge DGPS static surveys cost $1,362 per 5 points or $272 perpoint. Using conventional traversing methods would not have beenpractical for this project because the existing NCGS monuments weretoo far from the project area. McKim & Creed did compare levelingbetween the new GPS points, and the GPS elevation differences agreedwith leveling differences within 2 mm.

Highlights:

GPS is a practical and cost-effectivealternative to leveling for largeproject areas. The results farexceeded accuracy requirements forphotogrammetric control.

– CASE STUDIES


Lessons Learned

From the golf communityproject, McKim & Creedlearned the importance ofextensive mission planning andcontrol point location tooptimize the project control. Asa result, none of the GPS controlpoints were lost as a result ofconstruction activity. They alsolearned that more than two GPSreceivers are needed toefficiently perform a staticsurvey; they learned that fourGPS units appear to give themost value on a per-point basis.

From the bridge project, McKim & Creed learned that it needed toincorporate as many "known" vertical monuments as possible in order torefine the geoidal model. Otherwise, elevation differences would have beenlarger than 2 mm.

Highlights:

At $250 per point for GPS staticsurveys, GPS yielded savings of26% compared with the $340 perpoint costs of leveling.

Conventional surveys would havebeen impractical because of thelong distances from existingNCGS survey control. Elevationswere accurate between 0.2 and1.3-cm.


� ±�7,0(��&267�&203$5,6216

The projects described in this chapter were performed in accordancewith NGS’ (November 1997) guidelines for elevation surveys at the 2-cm accuracy level for ellipsoid heights, referenced previously.

5.1 Background

As part of the National HeightModernization Study, NGScompared leveling with GPSvertical surveys (using Geoid96)for determining NAVD 88orthometric heights; and evaluatedrelative differences in accuracy,time, costs, and other variables.Comparisons were performed inportions of California and NorthCarolina that have had difficultywith subsidence, crustal motion,and/or areas with largeuncertainties in the geoid model due to the lack of data. Surveylogbooks, and raw and processed GPS digital files were provided toNGS for accuracy comparisons. Time and cost data were also providedfor comparison.

Proposals were solicited from leading survey firms in California. CrossLand Surveying, Inc. of San Jose, California, and Johnson - Frank &Associates, Inc. of Anaheim, California, were chosen to survey two testareas selected by Don D’Onofrio, NGS’ California State Advisor.

Both firms were assigned different project areas along the Californiaaqueduct in Fresno County, west of the town of Mendota in California’sCentral Valley. Cross Land Surveying, Inc. was assigned the southernproject area, and Johnson - Frank & Associates, Inc. was assigned thenorthern project area. Both areas were approximately 30 kilometers in

Purpose:

• To compare leveling andGPS vertical surveys(with Geoid 96) fordetermining NAVD 88orthometric heights

• To evaluate relativedifferences in theaccuracies, time, costs,and other variables

– TIME & C OST COMPARISONS


length. Both firms were tasked to survey specified monuments using twoprocedures:

(1) Leveling using modified Second-order, Class I, double-runprocedures for 20 specified monuments at approximately 1-mileintervals

(2) GPS vertical surveys, using modified 2-cm guidelines, as detailedin the NOAA Technical Memorandum NOS NGS-58, datedNovember 1997, for 14 of the same monuments.

The purpose of these projects was to determine the relative differences incosts, accuracies, time, and other variables; therefore, certain cost-reduction compromises could be made that did not impact these relativecomparisons. Specifically, the leveling modification enabled single-runmodified double-simultaneous procedures to be used in certaincircumstances in lieu of double-run level loops; and the GPSmodification allowed GPS manufacturer’s software to be used in lieu ofNGS’ OMNI software.

Cost savings were highly variable for the following reasons:

• The contract surveys confirmed the significant cost savings anddesired accuracy level results found in the case studies, as well as theother additional benefits identified by the case studies. The costsavings are discussed in more detail below.

• In general, conventional surveying methods are more accurate thanGPS methods for height determination over relatively shortdistances. However, at the 10 kilometer spacing recommended forthe National Height System, GPS accuracies are comparable withthose attained by conventional surveying for most applications.

• One of the surveys highlighted an additional benefit of utilizing GPSto determine heights. A difference in the results obtained by the twomethods identified the need to refine the geoid model (used toconvert GPS-determined heights to those determined by conventionalmethods) for the local area. This benefit will be realized for otherareas if GPS methods are used to implement the National HeightSystem



5.2 California Southern Project

Cross Land Surveying, Inc., surveyedbetween benchmark PID GU4142(Z1444), at aqueduct post mile 111.91,and benchmark PID GU1348 (D1262),at aqueduct post mile 131.7. Theleveling was performed with a Wild

NA2000 electronic digital level, two 3-meter invar bar-coded level rods,and two Wild turning trivets. Of the total leveling distance, 27.5kilometers were surveyed with single-run leveling; and 9.6 kilometers(4.8 kilometers one way) were surveyed with double-run leveling. TheGPS surveys were performed with four Trimble 4000 SSE dualfrequency GPS receivers with L1/L2 geodetic antenna with groundplanes.

5.2.1 Accuracy Comparisons

The unadjusted heights are based on starting the level run at Point No.1348, with an elevation of 100.1877 meters, and running northwesterlyto Point No. 4142, utilizing field observed data only. The GPS elevationsare based upon a least squares adjustment utilizing Geoid 96 and holdingthe orthometric height of Point No. 1348 fixed at 100.1877 meters.

When utilizing the elevations from the “free” adjustment of the levelingdata, which held the elevation of Point No. 1348 at 100.1877 meters, thefollowing comparisons were made for the difference in orthometricheights between Point 1348 (southernmost point) and Point 4142(northernmost point):

(1) Height difference = 2.35665 meters from NGS’ 1989 unadjustedheights

(2) Height difference = 2.30296 meters from Cross’ 1998 “free”adjustment of leveling

(3) Height difference = 2.2911 meters from Cross’ 1998 rawunadjusted leveling heights (leveled heights with no adjustmentswhatsoever)

Reported by:Earl Cross, PLSCross Land Surveying, Inc.San Jose, CaliforniaTel: (408) 274-7994



(4) Height difference = 2.2784 meters from Cross’ 1998 “free”adjustment of GPS heights (Geoid 96)

The elevation of Point 1348 was assumed fixed at its 1989 elevation of100.1877 meters, although it probably subsided between 1989 and 1998.In relative comparisons between leveling and GPS surveys, the followingconclusions were reached:

• Comparison of (1) and (2) indicate that one of the two benchmarkssubsided over 5-cm more than the other benchmark between 1989 and1998.

• Comparison of (2) and (3) indicate the accuracy of the unadjusted rawlevel heights with regard to the minimally constrained height network.

• Comparison of (2) and (4) indicate potential orthometric heightdiscrepancies of only 2.46 cm, over the 30 kilometer project, between1998 leveling and GPS surveys.

The results proved that accuracies in the 2-cm range can be achieved, atthe 95% confidence level, utilizing GPS procedures, provided the Geoidmodel accurately represents the geoid in this area. The accuracy of theGeoid model is both the controlling and the limiting factor.

5.2.2 Time and Cost Data

Differential Leveling GPS Leveling Survey Phase

Staff Hours Costs Staff Hours Costs

Survey Planning – Labor 6.5 $488 12.5 $938

Deployment – Labor 12 (4x3) $1,050 12 (4x3 ) $1,050

Field Surveys – Labor 316 (4x79) $16,016 57 (4x13+5) $4,379

Survey Computations - Labor 19 $1,425 16 $1,181

Labor Subtotals 353.5 $18,979 97.5 $7,548

Other Direct Costs (ODCs) $2,661 $802

Totals $21,640 $8,350Table 12. California (South) Time/Cost Comparisons

Since the same 4-person survey team deployed for both surveys (GPSand leveling), the 12 man-hour deployments were not actuallyduplicated.



Reported by:Roger Frank, PLSJohnson-Frank & Associates, Inc.Anaheim, CaliforniaTel: (714) 777-8877

5.3 California Northern Project

Johnson - Frank & Associates, Inc. surveyedbetween benchmarks PID GU4142 and PIDGU0296. The leveling was performed withtwo Jena N1002 reversible compensatoroptical precision leveling systems. The GPSsurveys were performed with five Trimble4000 SSI/SSE dual frequency GPS receivers.

5.3.1 Accuracy Comparisons

The contract called for single run of the entire project and double run ofany section that did not match the NGS 1989 leveling by 6mm times thesquare root of the kilometers run. As the survey progressed, only one ortwo sections in the entire project would close with the record, while alldouble run levels would close within 1 or 2 mm on themselves. Afterdiscussion with the NGS California Advisor, it was decided that if theGPS matched the leveling, that would be a sufficient check, as it was notthe purpose of the project to determine the magnitude of subsidenceproblems in the area.

An unconstrained adjustment was made, holding only the mostsoutheasterly benchmark elevation, GU4142/Z1444. Statistically, leastsquares indicates a standard error in the total 30-kilometer project of 7.4mm. Holding only the most southeasterly benchmark, the survey missedthe most northwesterly benchmark, GU0296/92.58R, by 2.7 cm, asdetermined by 1989 raw leveling. Second Order Class I specificationsallow 3.3 cm in that distance. However, due to crustal movement,sectional misclosures with the 1989 NGS leveling at other bench marksalong the run were varied up to a maximum of 22 cm.

All GPS data were downloaded and baselines processed, using thebroadcast ephemeris, at the end of both days of GPS operation. After thesecond day’s baselines were completed, a check was made to ensure thatthe height element of the first day’s baselines matched those of thesecond day’s by less than 2 cm, per the NGS guidelines. One line wasfound to have a 2.5cm split. This line was re-observed, and the heightelement of the re-observed baseline was right in between the first two.

The Precise Ephemeris was downloaded from the NGS web site, whenavailable, and all GPS baselines were reprocessed. The resulting



baselines were imported in StarPlus’ StarNet, and an adjustment wasrun, holding only the most southeasterly benchmark and using Geoid 96to produce orthometric heights. Statistically, the standard error invertical for the 30 kilometers is 7.5 mm.

The orthometric height determined by GPS at the most northwesterlybenchmark was 16.4 cm lower than the height determined at the samepoint using leveling. In reviewing the intermediate benchmarks, it wasevident that the differences appeared to be on a slope from south tonorth, indicating a discrepancy between the height systems.

The heights derived from leveling, holding only the most southeasterlybench mark, were imported into StarNet, and held fixed to allow thesoftware to compute a best fit tilt in the north and east axes. Once thegeoid tilts were computed, the elevations were again set free except forthe fixed benchmark at the southeasterly end. Using this computed tilt inthe geoid, the GPS-derived orthometric heights then matched theleveling-derived orthometric heights to 1.2 cm or less in all locations.

Unlike the southern project, the orthometric heights derived from GPS differedfrom those derived from leveling by 16-cm. By using traditional leveling tohelp define and correct the local slope of the geoid, the orthometric heightsderived from GPS matched leveled heights to 1.2-cm or less at all locations. Asa side benefit, this project identified an inaccuracy in the geoid model in theproject area. This serves as an example of how implementing the NHS plannationwide will identify these types of differences in the geoid model and howthey will be resolved and incorporated into the NHS.



Reported by:Gary W. Thompson, RLSChief, North Carolina Geodetic Survey512 North Salisbury StreetRaleigh, North Carolina 27604-1118Tel: (919) [email protected]

5.3.2 Time and Cost Data

Leveling GPS Survey Phase

Staff Hours Costs Staff Hours Costs

Survey Planning – Labor 6 hr $574 15.8 hr $1,624

Deployment – Labor 27 hr $2,183 29 hr $2,708

Field Surveys – Labor 181.5 hr $12,908 77 hr $6,277

Survey Computations - Labor 23.5 hr $2,018 17.8 hr $1,695

Labor Subtotals 238 hr $17,683 139.6 hr $12,304

Other Direct Costs (ODCs) $1,740 $2,073

Totals $19,423 $14,377Table 13. California (North) Time/Cost Comparisons

Additional expenses were incurred by both survey firms for proposaland cost estimation, contract review and negotiations, administration,coordination with D&D and NGS, and report preparation.

5.4 North Carolina Project

The North CarolinaGeodetic Survey performedthe NHMS project in theAsheville area of westernNorth Carolina. Theproject extended from thedowntown area ofAsheville to the Eastern

Continental Divide, which is approximately 20 miles east of Asheville.The leveling route was 60 kilometers in length. The average differenceof elevation between sections was 14 meters, with the maximumdifference being -54 meters. The section length average was 0.75kilometers. The leveling was performed to Second Order Class Ispecifications. All new sections were double run. The leveling wasperformed with a Jena NI005A compensator optical precision levelingsystem with built-in micrometer and a Zeiss NI-2 compensator with anattached micrometer and four Kern GK-23E invar rods. NGS turningpins and thermistors were also used.

The GPS surveys were performed with four Trimble 4000SSE and twoTrimble 4000SSI dual frequency GPS receivers with L1/L2 geodetic



antennas with ground planes. Fixed height poles were used at all timesexcept at the Continuously Operating Reference Station (Base StationPID AA5552). The GPS data were processed with GPSurvey (Version2.3). The precise ephemeris was used to process the data. Theadjustment of the GPS data was performed with the NGS adjustmentprogram “ADJUST”. Geoid96 was used to obtain geoid heights. NCGSfollowed the procedures outlined in “Guidelines to Establishing GPS-Derived Ellipsoid Heights” (Version 4.3, 2CM Standard).

5.4.1 Comparison of GPS and Leveling

A free adjustment was performed holding one bench mark (E 39, PIDFB0803, First Order Class I) and one HARN (K 180 PID FB0035) fixed.The elevations obtained from this adjustment were compared to thepublished elevations of bench marks occupied with GPS and with theadjusted elevations obtained from the leveling performed in this project.The average difference between the GPS and leveling orthometricheights was -0.015 meters with the largest difference being -0.031meters. The largest differences occurred in the eastern area of theproject near the Eastern Continental Divide.

The results of this project indicate that 2-5 centimeter heights can beobtained at the 95% confidence level, utilizing proper field proceduresand a good geoid model.

5.4.2 Time Comparison (GPS versus Leveling)

The time comparison did not include the staff hours for reconnaissancefor the GPS or leveling phase. Mark recovery and mark setting isrequired to perform both GPS and leveling. The reconnaissance for GPSdiffers slightly, but statistically they are equal. Also, additional geodeticmarks were recovered along the level route. Consistent with standardpractice of the NCGS, additional marks along the level route werepositioned vertically. Positioning these additional marks in the levelingphase did not affect the comparison of staff hours between geodeticleveling and the GPS observations.



Field Survey Phase(Staff hours)

Leveling (2nd

Order ClassI)

GPS (2cmStandard)

Field Observations 1,111 282

Computations 25 25

Totals 1,136 307

Table 14. North Carolina Geodetic Survey Time Comparisons

Using the time comparison above, the cost to perform the NorthCarolina project using geodetic leveling techniques would increaseby 270% when compared with the cost of performing the projectusing GPS

5.4.3 Project Statistics

GPS Elevation Surveys

Total number of stations occupied = 39

Existing horizontal stations = 11

Existing vertical stations = 3

Existing horizontal/vertical stations = 12

GPS stations established = 13

Leveling

Total number of stations occupied = 81

Existing vertical stations = 41

New vertical stations = 40



5.5 Summary

The time and cost data for the three projects are summarized in Table 15.

Leveling GPS

Km Time Cost Points Time Cost

S. Calif. 30 354 $21,640 14 97 $8,350

N. Calif. 30 238 $19,423 14 140 $14,377

NorthCarolina

60 1,136 Costs notAvailable

39 307 Costs notAvailable

Averages 14.4 staff hours/kmleveled

$684/km leveled (flatterrain)

8.1 staff hours/GPS point

$811/GPS point

Table 15 Statistics from Cost Comparison Projects

Costs of GPS elevation surveysare estimated by the number ofpoints surveyed. Within reason,distance is not a factor. WithGPS, surveying a point 5 milesaway costs very little more thansurveying a point 2 miles awayfrom the DGPS base station.

Relative to the other projects, the GPS costs were higher in the NorthCalifornia project because of unexpected discrepancies betweenorthometric heights derived by GPS compared with those from leveling.

Cross Land Surveying, Inc. costs averaged approximately $600 per GPSpoint, whereas Johnson-Frank, Inc. costs averaged over $1,000 per GPSpoint. Cross encountered no major difficulty with the Geoid 96 model inthe southern area, whereas Geoid 96 appeared to inadequately representthe geoid in the northern area surveyed by Johnson-Frank & Associates.Subsequently, Johnson-Frank resurveyed some areas, extensivelyanalyzed and remodeled the geoid from leveling, and incurred additional

GPS costs averaged $811 perpoint. Leveling costs averaged$684 per kilometer. The levelingcosts are very conservative, basedon the two California projects inflat terrain.



GPS costs from equipment rental. These were the major reasons whytheir costs per GPS point were higher.

The costs for leveling areprimarily estimated by thedistance surveyed. Withinreason, the number of pointssurveyed is not a factor.Surveying 20 points along a 1-mile path costs very little morethan surveying only 1 or 2points along that path. In fact,Johnson - Frank, Inc., leveled49 bench marks rather than the20 set forth in the contract;leveling through all availablebenchmarks costs no morethan leveling around them. .

Costs for leveling can and willincrease significantly in hillyor mountainous terrain. Theaverage leveling costs were$684.38 per kilometer in flatterrain, and $1,352.22 perkilometer in hilly terrain.Here, GPS again has theadvantage. It makes little difference in GPS surveying whether thebaseline is level or extends into the mountains, other than the travel timerequired between the two ends of the baseline.

Adverse weather impacts both leveling and GPS, but in different ways.Leveling is not performed in the rain, and both leveling crews weredelayed two days because of rain. GPS operations can proceed in therain; but during these demonstration projects, one GPS session wasstopped as a weather front was passing through the area. NGS’guidelines require meteorological data (normally, wet- and dry-bulbtemperatures, and atmospheric pressure) to be collected regularly andespecially immediately before and after an obvious weather front passesduring a session, if possible. Atmospheric pressure measurements mustbe made at approximately the same height as the GPS antenna phasecenter. Even though these data may not be used in the vector processing,

Highlights:

For the Southern Californiaproject, GPS costs were 39% thoseof leveling, and accuracy wascomparable.

For the Northern Californiaproject, GPS costs were 74% thoseof leveling, and accuracies weredebatable.

For the North Carolina project,cost data were not available, butGPS times were only 27% those ofleveling, and accuracy wascomparable.

In general, leveling costs aredistance based, whereas GPS costsare point based. GPS is clearly thekey in linking the NHS networknationwide.



• The GPS network will have superior relative and absolute accuracies of 3-Dcoordinates. Additional GPS receivers form a stronger network ofsimultaneous baselines, and multiple GPS surveys relative to an assumed basestation cause that station to approach the accuracy of a 3-D HARN point.

• With network point spacing of 10 Km, the estimated cost savings are 88%, andthe estimated time savings are 94%.

they may be helpful duringthe analysis of the resultsand in future reprocessingwith more robust software.

From an accuracyperspective, levelingextends orthometric heightsmore accurately from onepoint to the next; and GPSextends ellipsoid heightsmore accurately from onepoint to the next. For GPS-derived orthometric heightsto approach the accuracyderived from leveling, thegeoid model needs to be asaccurate as possible.Therefore, any effort toimprove and evaluate theaccuracy of NGS’ geoidmodel, nationwide, servesto directly improve theutility of GPS nationwide.

The next four pages address the expansion of surveys from singlebaselines into complex networks. Although five (5) GPS receivers areused in the following examples (GPS base station plus four “rovers”),efficiencies will be further improved as additional GPS receivers areused for simultaneous observation of multiple baselines.

The northern GPS demonstrationproject was essentially a worst-case scenario. Differencesbetween the height systems wereevident. These differences needto be evaluated. If GPS alonehad been used, no one wouldhave been aware that theorthometric heights were poor.It took a combination of GPS,leveling, and NAVD 88 heights tobring out the best in each. Thiswas the major lesson learnedfrom these projects.

Until NGS’ geoid model isevaluated nationwide, every GPSsurveyor will be uncertain as tothe accuracy of GPS-derivedorthometric heights. NGS’ geoidmodel cannot be sufficientlyevaluated nationwide until theNHS is implemented.



5.5.1 Cost Comparisons – Single Baselines

For two points connected by a single baseline with lengths between 1and 10 kilometers, the comparative costs between leveling and GPS areshown in Table 16. The results are plotted by the graph at Figure 16.

Single BaselineLength

LevelingCosts

GPSCosts

GPSSavings

1 Km $680 $1,620 -138%

2 Km $1,360 $1,620 -19%

3 Km $2,040 $1,620 21%

4 Km $2,720 $1,620 40%

5 Km $3,400 $1,620 52%

10 Km $6,800 $1,620 76%

Table 16 Comparative Costs for Single Baselines

Figure 16 GPS Cost Savings (%) for Single Baselines

-140

-120

-100

-80

-60

-40

-20

0

20

40

60

80

100

1 2 3 4 5 6 7 8 9 10

Single Baseline Length (Km)

Co

st S

avin

gs

(%)

Usi

ng

GP

SBASELINE



Cost Comparison – Multiple (Network) Baselines

For an expanding network with four new points at a time formingmultiple baselines, with each baseline length between 1 and 10kilometers, the comparative costsbetween leveling and GPS areshown in Table17. The results areplotted by the graph at Figure 17.

NetworkBaselineLength

TotalLevelingLength

LevelingCosts

GPSCosts

GPSSavings

1 Km 4 Km $2,720 $3,240 -19%

2 Km 8 Km $5,440 $3,240 40%

3 Km 12 Km $8,160 $3,240 60%

4 Km 16 Km $10,880 $3,240 70%

5 Km 20 Km $13,600 $3,240 76%

10 Km 40 Km $27,200 $3,240 88%

Table17 Comparative Costs for Multiple (Network) Baselines

Figure 17 GPS Cost Savings (%) from Multiple (Network) Baselines

-20

-10

0

10

20

30

40

50

60

70

80

90

100

1 2 3 4 5 6 7 8 9 10

Network Baseline Length (Km)

Co

st S

avin

gs

(%)

Usi

ng

GP

S

BASELINE

4-POINT BOX



5.5.2 Time Comparisons – Single Baselines

For two points connected by a single baseline with lengths between 1and 10 kilometers, the comparative survey times between leveling andGPS are shown in Table 18. The results are plotted by the graph atFigure 18.

Single BaselineLength

LevelingStaff Hours

GPSStaff Hours

GPSSavings

1 Km 12.8 16.6 -29%

2 Km 25.6 16.6 35%

3 Km 38.4 16.6 57%

4 Km 51.2 16.6 66%

5 Km 64.0 16.6 74%

10 Km 128.0 16.6 87%

Table 18 Comparative Times for Single Baselines

Figure 18 GPS Time Savings (%) for Single Baselines

-30-20-10

0102030405060708090

100

1 2 3 4 5 6 7 8 9 10

Single Baseline Length (Km)

Tim

e S

avin

gs

(%)

Usi

ng

GP

S

BASELINE



Time Comparison – Multiple (Network) Baselines

For an expanding network with four new points at a time formingmultiple baselines, with each baseline length between 1 and 10kilometers, the comparative timesbetween leveling and GPS are shownin Table19. The results are plottedby the graph at Figure 19.

NetworkBaselineLength

TotalLevelingLength

LevelingStaff

Hours

GPSStaff

Hours

GPSSavings

1 Km 4 Km 51.2 33.2 35%

2 Km 8 Km 102.4 33.2 68%

3 Km 12 Km 153.6 33.2 78%

4 Km 16 Km 204.8 33.2 84%

5 Km 20 Km 256.0 33.2 87%

10 Km 40 Km 512.0 33.2 94%

Table 19 Comparative Times for Multiple (Network) Baselines

BASELINE

4-POINT BOX

0

10

20

30

40

50

60

70

80

90

100

1 2 3 4 5 6 7 8 9 10

Network Baseline Length (Km)

Tim

e S

avin

gs

(%)

Usi

ng

GP

S

Figure 19 GPS Time Savings (%) for Multiple (Network) Baselines



The leveling times reflect both the impact of leveling over hilly terrain inNorth Carolina and the flat terrain in both California projects. On theother hand, GPS is terrain independent.

Although GPS averaged 32% more expensive per hour than leveling, itproved far more efficient and quickly becomes more cost effective asdisplayed in the tables and graphs above. The actual costs of the twoCalifornia projects averaged $71.42/hour for leveling and computations,and $94.54/hour for GPS surveys and computations. These were costsactually incurred, rather than contracted costs based on a priori estimatesand known labor rates.


� ��),1',1*6

Five major findings were distilled from users needs assessments,presented at the user forums, and follow-up discussions andclarifications from users in principal application categories. Wherepossible, cost-benefits are provided.

6.1 Finding #1. America needs a “reliable” and efficient meansto determine “absolute” elevations, relative only to theearth’s center.

• The word “reliable” is in quotes because America (and the world)needs GPS to be secure from jamming, free from Selective Availabilitythat deliberately degrades the GPS signals, and upgraded withadditional civil frequencies to improve GPS performance in cities,forests, and other places with obstructed vision of the sky.

• The word “absolute” is in quotes because most elevations in use todayhave “relative accuracy” rather than “absolute accuracy.” Benchmarkelevations are: (1) relative to the accuracy of other benchmarks fromwhich leveling was performed, (2) relative to survey procedures used,(3) relative to the distance surveyed inland from “mean sea level,” and(4) relative to the route surveyed to get to the current benchmark. It isvirtually meaningless to say that a benchmark is accurate to 2-cm (atthe 95% or other confidence level), for example, because levelingaccuracies refer to X parts per million or similar relative measurementstatistics. The same is not true of GPS, when surveys are referenced toCORS stations.

• NGS already operates many CORS stations that are surveyed so wellthey are assumed to have zero errors relative to the earth’s center. Thismeans that all GPS surveys relative to the CORS essentially have“absolute accuracy.”

• NGS already publishes guidelines for GPS elevation surveys at the 2-cm and 5-cm levels relative to the CORS.

-- FINDINGS


• NGS already publishes a Geoid model that enables users to convertfrom elllipsoid heights (from GPS) to orthometric heights (fromleveling). When the Geoid model is further refined, to 1-cm accuracynationwide, users will be able to reliably survey orthometric heightswith accuracies on the order of one inch, relative to the earth’s center.This is the ultimate goal of many users.

• This requirement is essentially do-able now, provided resources areavailable to convert and expand the current (2-D) High AccuracyReference Network (HARN) into a 3-D network.

6.2 Finding #2. Based on DGPS and CORS, America needsnationwide implementation of a standardized verticalreference datum as the legal basis for elevation data.

This finding, too, is technically do-able now, but the neededinfrastructure is not fully in place toimplement NAVD 88 as describedherein. .Figure 24 projects anaverage growth in GPS usage of$1.49 billion annually for ten years.

Users are universally confused bythe bewildering array of verticaldatums in use today throughout theUnited States, many of which are not linked to NAVD 88 or NGVD 29..

Implementation of NAVD 88 is a prerequisite for satisfaction of theremaining modern height requirements.

6.3 Finding #3. America needs high accuracy, high resolutionDigital Elevation Models (DEMs) based on NAVD 88.

• Although they are the most cost-effective DEMs available to thepresent time, DEMs available from USGS have three majorlimitations: (1) their elevation accuracy is relatively poor (root meansquare errors of 7 meters for Level 1 DEMs; 10 feet for Level 2

Users want to know theelevation of a point, andnot be confused by thefact that the elevation isX-feet when using NAVD88, Y-feet when usingNGVD 29, Z-feet whenusing IGLD85, etc.

-- FINDINGS


DEMs); (2) their point spacing is too wide (typically 10 to 30 meters);and (3) they are produced to the NGVD 29 datum.

• LIDAR and/or IFSAR DEMs solve all three of these problems, but at aprice. Although cost estimates vary, it currently costs USGS anestimated $75 per square mile for photogrammetric contouring of quadmaps, additional costs for production of hypsography and hydrographyDLGs if required, plus $10 per square mile to produce the Level 2DEMs. When mass-produced, the most accurate DEMs producedfrom LIDAR would cost an estimated $500 per square mile.

• As a less expensive alternative, IFSAR-generated DEMs (1-3 meteraccuracy) could be produced nationwide for an estimated $100 million,i.e., as low as $25 per square mile when mass-produced.

• In five years using IFSAR and/or LIDAR, America could have DEMswith elevation accuracy measured in inches or feet, rather than meters.The DEM point spacing would be 5- to 10 meters, rather than 10 to 30meters. The vertical datum would be NAVD 88, rather than NGVD29. The DEMs would be so good that contour lines could be generatedat any desired contour interval, bypassing conventional lengthy andcostly photogrammetric contouring; and these DEMs would supportthe production of larger-scale digital orthophotos. With LIDAR and/orIFSAR DEMs, virtually every American would benefit from the data,and the benefit-cost (B/C) ratio would be large, benefiting virtuallyevery user application evaluated in this study.

Areas with DEMApplications

Estimated Valueto Constituents


Nationwide Terrain $33.5 million • Replace less-accurate Level 1 DEMsthat cost USGS approximately $33.5Million

• Provide 6" DEMs costing $500 onaverage per mi2, in lieu of 1’contours costing $5,000 per mi2

• Enable rapid generation of contoursfor USGS maps and GISsnationwide

• Enable 3-D modeling by USACE,FHA, FRA, FAA, EPA, USFS, etc.

Nationwide Watersheds $100 million • Automated hydrologic modeling byNWS and FEMA to predictlocations/ volumes of peak waterconcentrations

-- FINDINGS


Special Flood HazardAreas (SFHAs)

$225+ million • Automated hydraulic modeling byFEMA to determine depth andextent of flood waters

• Determination of flood risks andinsurance rates

Coastal Erosion Zones $11.25+ million • Accurate determination of coastalerosion rates

• Determination of insurance ratesUrban Areas $500 million • Urban planning

• Intelligent Transportation System(ITS) planning

• Elevation layer in GIS database• Stormwater management

Farm Lands $1.7 billion • Precision farming for plannedapplication of water, fertilizer, etc.

• Control of unwanted run-off andstream contamination

Forest Areas Not estimated • Quantification of timber volumes• Models for spread of wildfires• Plans for wildfire mitigation

HAZMAT Areas Not estimated • Analyses of contaminated sites• Plans for clean-up

Totals $2.5+ billion

Table 20. High Accuracy DEM Beneficiaries

6.4 Finding #4. Based on NAVD 88, America needs aNationwide Differential GPS (NDGPS) system for real-timeaccurate 3-D positioning; for navigation, tracking, publicsafety, precision farming, and construction

• America needs the NDGPS proposed by the Department ofTransportation, and championed by the U.S. Coast Guard (USCG) andthe Federal Railroad Administration (FRA). Using surplus Air Forcecommunications equipment, the NDGPS pilot project in Appleton,Washington is working perfectly. Every American who drives amodern vehicle or uses modern telecommunications in the future, andevery farmer who uses precision farming, would benefit from thisinitiative. The B/C ratio is several hundred to one, when justifiedsolely on the basis of horizontal positioning benefits to America. TheB/C ratio improves significantly when its 4-dimensional benefits(latitude, longitude, elevation, and time) are taken into account. The15-year life-cycle costs for NDGPS are only about $70 million, and thebenefits to America are in excess of $100 billion.

-- FINDINGS


NDGPS Applications Estimated Valueto Constituents


Vehicle Positioning andSafety

$8.385 billion • Automated VehicleLocation (AVL)

• Computer-AidedDispatching (CAD)

• Intelligent TransportationSystem (ITS)

Train Location and Safety $67.34 million • Positive Train Location(PTL)

• Positive Train Separation(PTS)

Maritime Navigation andSafety

$9.6 billion • Positioning of dredges• Positioning of cargo

shipsMining and HeavyConstruction

$90 billion • Computer-AidedEarthmoving System(CAES), estimated 10%of total savings fromGPS modernization

• Real-time control ofequipment

Precision Farming $1.803 billion • Real-time control of farmequipment

• Controlled application ofwater, fertilizers,pesticides, etc.

Forest Management $6.83 million • Real-time positioning

Environmental Protection $6.33 million • Real-time positioningand damage assessments

Disaster Response $7.5 million • Real-time positioningand damage assessments

• Geocoded addressingwhen street signs andnormal address systemfails

Surveying Industry Not estimated • Vastly improved surveyprocedures

State/Local Governments $178.05 million • AVL and CAD for policeand other E-911 vehicles

• Infrastructure surveysand management

Totals $110.052 billion Over the 15-year life-cycleof NDGPS

Table 21. NDGPS Beneficiaries

-- FINDINGS


6.5 Finding #5. America needs cost-effective, mass-producedGPS elevation surveys of all buildings in or near floodplains,as well as coastal areas vulnerable to hurricane tidal surges,that are based on NAVD 88.

America needs accurate GPS elevation surveys, mass-produced for anestimated 10,000,000 flood- and hurricane-prone buildings in or nearknown flood hazards, in order to: (1) implement proactive floodplainmanagement, (2) resolve uncertainties as to flood risk, and (3) get ownersof floodprone buildings to purchase needed flood insurance. FEMAknows that horizontal criteria, used by the mortgage industry for flood riskdeterminations, needs to be replaced with accurate elevation surveys andvertical criteria; NHS implementation proposed here will help solve this.Although this requirement carries a nationwide “price tag” ofapproximately $1 billion, elevation surveys would pay for themselves inone year if only half the owners of flooded buildings had been convincedto purchase flood insurance. This would reduce dependence on Federal“bail-outs” ($2+ billion per year) when inevitable floods occur.

Figure 20 GPS Elevation Certificate (Pre-Flood)

-- FINDINGS


Beneficiaries of this initiative would include the following, all of whomneed to know the true flood risk of individual buildings: (1) potentialowners, (2) real estate industry, (3) mortgage industry, (4) insuranceindustry, (5) construction industry, and (6) local government officials.

When hurricanes are accompanied by tidal surges, FEMA needs toquickly survey the breadth and depth of tidal surges, and the elevation ofindividual damaged buildings, to determine those eligible for homeowner insurance reimbursement (wind damages only). Most homeowner policies do not cover floods or hurricane tidal surges. Suchelevation surveys would be proactive if performed prior to naturaldisasters, and used to mitigate potential losses.

Strong arguments can be made that Flood Insurance Rate Maps(FIRMs) would not be required if GPS Elevation Certificates wereavailable for all buildings in or near floodplains. Flood InsuranceStudies (FIS) would still be required to the point where Base FloodElevations (BFEs) are computed. Whereas the Special FloodHazard Area (SFHA) boundaries on FIRMs (the horizontal criteria)indicate broad areas vulnerable to flooding, the GPS ElevationCertificates indicate actual depths of interior flooding, for individualbuildings, for the 100-year (1% annual chance) base flood, based onBFEs (the vertical criteria). Thus, specific flood risks would beaccurately determined for individual buildings (suitable for floodinsurance actuarial rate determinations) based on vertical criteria,as opposed to generalized flood risks for all buildings in SFHAs,based on horizontal criteria. ACCURATE VERTICAL CRITERIACOULD REPLACE CONTROVERSIAL HORIZONTAL CRITERIAIN FLOOD RISK DETERMINATIONS, SOLVING A MAJORPROBLEM CONFRONTING 15 MILLION NEW HOMEOWNERSANNUALLY IN THE UNITED STATES.

-- FINDINGS


North Carolina leads the nation in terms of “proactivefloodplain management.” Floods are our most predictablenatural hazard, and GPS Elevation Certificates, such as shownin Figure 20, certify the elevation of the lowest adjacent grade,the lowest floor, and the base flood elevation (BFE), from whichthe depth of interior flooding is computed for the 1% annualchance flood. This is the most understandable and authoritativedocument for flood hazard/risk assessment – the first step of“flood mitigation.” It is not true that “floods happen, andthere’s not much we can do about it.” There are in fact manyproactive steps that can be taken to reduce the risk of potentialfloods, and reduce the pain and loss when floods occur.


� ��5(&200(1'$7,216

7.1 Summary

As has been discussed in this report, the North American Vertical Datumof 1988 (NAVD 88) is the practical realization on the ground of asophisticated elevation reference system for the North Americancontinent. It is designed to be integrated into a seamless network ofhorizontal and vertical reference points, gravity data, GPS satellites, andtracking stations. This proposed National Spatial Reference System(NSRS), with NAVD 88 as its elevation reference, would support suchdiversified uses as:

• Precise navigation and aircraft landing systems

• Floodplain management and the National Flood InsuranceProgram

• Highway and railroad transportation infrastructure

• Intelligent vehicle highway systems

• Earthquake, volcanic, and subsidence research programs

• Disaster preparedness and relief efforts

• Water supply and delivery infrastructure


• International boundaries and offshore boundary mapping

• Coastal zone management

• Environmental cleanup and ground water monitoring

While NAVD 88 was designed and executed over seven years ago, itsimplementation into the NSRS has yet to be achieved, and itsdeficiencies in certain regions threaten its very existence. In California,only 30% of the existing vertical control monuments were ever includedin NAVD 88 because of their uncertain stability. Of this network of17,000 NAVD 88 monuments, subsidence or seismic activity has now

-- RECOMMENDATIONS


significantly disturbed 25%. Similar subsidence issues exist in Texas,Louisiana, Florida, South Carolina, North Carolina, Kentucky, Ohio,Virginia, New Jersey, New York, Michigan, and Minnesota. TheNational Academy of Sciences has estimated subsidence damage costs ineach of these states to exceed $10 million annually. Relatively smallchanges in elevation often have profound impact. For instance:

• The National Research Council conservatively estimated the costsresulting from increased flooding and structural damage fromsubsidence in the United States to be in excess of $125 million peryear.

• The California Department of Water Resources estimates aquifers inthe state are over drafted by 10 million acre-feet annually—anincrease of 75% over the last five years.

• The National Science Foundation in conjunction with JPL, NASA,Scripps Institute, UCLA, and USGS provided $7.5 million infunding in 1996 for GPS earthquake research.

The use of the 750,000 precisely located, in-the-ground or monumentedreference points installed over the past 200 years to measure heights isnot adequate to meet the needs of today's mobile and technology-drivensociety. The classical "line-of-sight" measurements do not provide thereal-time accuracy needed for today's positioning technologies andapplications, including precision agriculture, efficient marinetransportation, and zero visibility landings of aircraft. In addition, manyof these reference points have been disturbed, destroyed, or are not incompliance with today's requirements for accuracy.

The implementation of NAVD 88 means densifying the network throughrecomputation of existing data and execution of new surveys and studiesto bring the horizontal, vertical, and gravity control networks togetherinto a unified system joined and maintained by GPS. US Code, Title 33,Section 883 and the Office of Management and Budget Circular A-16(revised October 19, 1990) specify this authority and responsibility to bethe mission of the National Geodetic Survey.

NGS has been unable to apply appropriate resources to undertake acomprehensive solution to the challenge. What is needed is a Nation-wide effort for height modernization led by pilot or demonstrationprojects in the most difficult and time-critical areas of crustal motion

-- RECOMMENDATIONS


under the direction and expertise of NGS. This is fully compatible withthe agency’s official charge and its recently published mission, visionand Goals. New surveys and studies would be accomplished undercontracts to private firms promoting the goals of economic benefit andtechnology transfer. Development of standards, contracting oversight,technical supervision, final analysis, and publication would be conductedby NGS ensuring consistency and complying with existing locallegislation. In its finished state, the NSRS and NAVD 88 would providea consistent three-dimensional framework for positioning throughoutNorth America that is fully compatible and maintainable by GPS and/orother space-based navigation systems. The combination of an improvednational height system (North American Vertical Datum of 1988–NAVD88) first adopted by the Federal government in 1993, with thepositioning technology of the GPS, offers the nation and itsgovernments, for the first time, the ability to obtain precise verticalmeasurements in real-time.

7.2 Desired Outcome/Objective

The most desirable outcome is a unified national positioning system,comprised of consistent, accurate, and timely horizontal, vertical, andgravity control networks, joined and maintained by GPS andadministered by the National Geodetic Survey (NGS).

A state-of-the-art National Spatial Reference System (NSRS) withNAVD 88 as its elevation reference can make available to the nation acommon, consistent set of real-time geographical coordinates orreference points. The applications of this break-through nationalpositioning system will provide:

• Improved coastal and harbor navigation allowing for greatercost-effective transshipment of goods,

• Advanced surface transportation control and monitoring,

• Production of accurate Digital Elevation Models (DEM)allowing for better floodplain analysis and flood insurance needs,

• Highly efficient fertilizer and pesticide spreading, resulting inreduced run-off water pollution and more competitive farmingthrough lower costs and higher crop yields,

-- RECOMMENDATIONS


• More accurate modeling of storm surge and pollutiontrajectories,

• Better monitoring of crustal movement to allow for improvedunderstanding of tectonic movement and improved earthquakeresistant designs,

• Increased reliability for improved resource management decisionmaking though the use of Geographic Information Systems,

• Support the modernization of our transportation infrastructureand the mission and goals of the Intermodal SurfaceTransportation and Efficiency Act (ISTEA),

• Improved airline and aircraft safety through GPS controlledapproach and landing, and

• Accurate and consistent vertical data for building major crossborder projects with Canada and Mexico in support of America’senvironmental objectives.

7.3 Recommendations for Implementing NAVD 88

7.3.1 Approach

To implement NAVD 88 throughout the United States, an approach isrecommended that involves both Federal and private sector forces.Realizing that this implementation will carry a high short-term cost andan ongoing implementation cost, a two-phased approach isrecommended. The first phase would focus on the survey work neededto establish the NAVD 88 reference bench marks, while the secondphase would expand the system to more users.

Phase 1

Phase one would be a cooperative effort between Federal and privatesector forces to establish the Federal Base System (FBS) a nationwidenetwork of 3-D control monuments with 10-kilometer spacing, over arecommended period of 5 years. It would include:

(1) Geodetic surveying; activities associated with the projects,including identification of monuments to determine whichmonuments are suitable for GPS occupation in the project areas;performing leveling and collecting GPS observations; processing

-- RECOMMENDATIONS


and adjusting GPS and leveling data; and preparing reports andsubmitting results for publication.

(2) Development of a capability within the NGS to manage theimplementation effort, determine priority areas, select qualifiedprivate sector contractors, manage contracts, and providetechnical supervision.

(3) Analyze and document the accuracy of the heights obtained bythe projects.

(4) Publish GPS-derived ellipsoid and orthometric heights with theirassociated accuracy.

(5) Perform technology transfer activities, including publishing non-technical documents describing NAVD 88 and GPS heights,

(6) Conducting seminars to promote understanding of NAVD 88 andGPS for non-technical persons and technical personnel,

(7) Developing training seminars for technical persons to becomeinstructors in the use of GPS to implement NAVD 88, and

(8) Presenting workshops to train the private sector to properlyprocess and submit GPS and leveling data to NGS

It is recommended to use private sector firms to provide the necessaryfield and GPS survey work to survey the FBS. All private sector workwould be out-sourced on a project basis in accordance with QualificationBased Selection (QBS) procedures and the Federal AcquisitionRegulations (FAR). All contacting would be carried under the directsupervision of NGS.

Phase 2

Phase two would be the ongoing expansion of the Federal Base System,and it should be carried out under the direction of the NGS. Themaintenance effort would be funded on a yearly basis and would involvethe cooperation of the states. This is similar in nature to the methodsused to maintain the horizontal reference systems in place today. Itwould involve the expertise of the NGS through their State Advisorprogram.

-- RECOMMENDATIONS


State-by-State Implementation

The most cost effective and manageable method to implement theFederal Base System for NAVD 88 throughout the continental UnitedStates will be on a state-by-state basis. It will be the responsibility of theNGS to set the priorities for each state, based on the following criteria:

• Amount of seismic activity and subsidence within the state,

• Degree of urbanization and development,

• Needs of the user community, and

• Support for the program within the state.

Once the priorities have been established, the NGS will develop a workplan consisting of a technical scope of work, schedule, and budget foreach target state. This work plan will be used to set the level of fundingfor the term in which the implementation will be carried out.

Demonstration Projects

Two of the states having the greatest immediate need for the fullimplementation of NAVD 88 are California and North Carolina. Bothstates are subject to extreme seismic activity, subsidence, floodplainmanagement, coastal erosion, and heavy urbanization. With this report,it is recommended that two demonstration projects be undertakensimultaneously at the inception of phase one. These demonstrationprojects would accomplish the following mission and be completedduring the first year of the program:

• Develop the needed program management capabilities within NGS.This will allow the NGS to put in place the necessary internalprocedures and staffing to manage this effort;

• Allow the NGS to develop the proper Federal procedures for out-sourcing of field survey and technical services;

• Train internal NGS staff in the requirements for contract managementand technical supervision;

• Test the ability of private sector contractors to carry out this mission;

-- RECOMMENDATIONS


• Refine technical requirements and standards for using GPS forproviding the required degrees of precision;

• Develop a more precise cost model for the implementation of NAVD88 in the remaining states; and

• Provide a status report, to the House, detailing the program’s progress,recommendations for the next phase and level of funding, benefitsincurred, and lessons learned.

7.3.2 Time Frame

The sooner the Federal Base System for NAVD 88 is fully implemented,the sooner the Nation will be able to reap the benefits. The programcould be implemented over a period of five to ten years depending uponthe level of funding available. Using a combination of Federal andprivate sector forces, this time frame is entirely feasible. However, thiscannot happen without additional resources as described below.

7.3.3 Costs

Tables 22 and 23 show the estimated costs for implementing the FederalBase System, including NAVD 88, at a basic 10-kilometer spacing. Thisincludes all costs for providing new field surveys and oversight by theNGS. Table 22 indicates the estimated costs for the first phasedemonstration projects, and Table 23 includes costs for nationwideimplementation.

7.3.4 Methodology

To best serve the user community, a Federal Base Network isrecommended, based on a nominal 10-kilometer spacing of permanentNAVD 88 reference points. This will allow for the maximum effectiveuse by the highest percentage of users. A vast majority of the privatesurvey firms in the United States are small businesses, with staffs under10 persons. These firms generally cannot afford the cost for advancedGPS technology, the type needed to produce precise and accurateheights. Until the cost for this GPS technology reaches an affordablelevel, these firms will rely on the use of leveling to serve the heightneeds of their communities and clients. A denser, 5-kilometer spacing,while recommended in the urban areas, would cost almost four times asmuch as the 10-kilometer spacing.

-- RECOMMENDATIONS


Nominal spacing of 10- kilometers would yield approximately 55,000permanent NAVD 88 reference points throughout the continental UnitedStates, (eliminating points where monuments are obviously not required)at an average cost of approximately $1,200 per 3-D survey monument.The $1,200 estimate is based on the approximate $800 average cost ofobserving with GPS plus an estimated $400 required to cover the point’sreconnaissance, monumentation, and administrative costs. This wouldallow the local surveyor or engineer easy access to the system. In nocase would they have to travel more than 5 kilometers (3.1 miles) to gainaccess to the system. This would allow for more cost-effective surveysand encourage the use of the system. Spacing greater than 10-kilometerswill have a detrimental effect to both of these issues, and defeat many ofthe benefits described in this report.

Table 22 Phase 1 – First Year Demonstration Projects

Activity Estimated Cost of UsingConventional Surveying

Technologies

Estimated Cost ofUsing GPS

Technologies

State California NorthCarolina

California NorthCarolina

Subtotal $41,200,000 $20,040,000 $4,600,000 $2,380,000

TOTAL $61,240,000 $6,980,000

Table 23 Nationwide Implementation of the Federal Base System

Activity Estimated Costs ofUsing Conventional

Surveying Technologies

Estimated Costs ofUsing GPS

Technologies

TOTAL $596,000,000 $66,000,000


� ��$33(1',;

This Appendix provides detailed technical background to clarifydiscussion in the remaining chapters of the report

8.1 National Spatial Data Infrastructure (NSDI)

Information about where an object or feature is or where an event takesplace often is an important factor in decision making in both the publicand private sectors. Geospatial data, which identify the geographiclocation and characteristics of natural or constructed features andboundaries referenced to the earth, provide a unique context forintegrating otherwise disparate observations and for evaluatingcompeting options. Factors of location, distance, pathways, and otherspatial relations often must be considered when making decisions abouteconomic ventures, resources management, environmental and healthconcerns, and responses to emergencies.

Public and private sector organizations have recognized the usefulness ofspatial data in their activities. The U.S. spends billions of dollarsannually on the collection, management, and dissemination of spatialdata. Advances in computer techniques to collect and process spatialdata, together with decreasing costs for acquiring these technologies,help organizations using spatial data to do so more efficiently andeffectively. Such advances enable other organizations to use spatial datafor the first time. Technologies such as the Internet and the World WideWeb enable organizations to make their information more widelyavailable and to locate data produced by others.

The NSDI facilitates data sharing by organizing and providing astructure of relationships between producers and users of spatial data.By participating in the NSDI, Federal, state, regional, and localgovernment agencies; companies; and nonprofit organizations cancooperate to develop consistent, reliable means to share spatial data.Executive Order 12906, "Coordinating Geographic Data Acquisition andAccess; the National Spatial Data Infrastructure," dated April 11, 1994,formalized Federal participation in initial efforts to implement the NSDI.

APPENDIX


Instructions in this executive order are that Federal agencies will workwith non-Federal organizations to develop the NSDI, will document theirspatial data and make this documentation available to the public, andwill make plans to provide public access to their spatial data. Thisexecutive order also instructs agencies to lead in the development ofstandards.

Office of Management and Budget (OMB) Circular A-16 established theFederal Geographic Data Committee (FGDC) to develop the NSDI. Thiscircular assigns to Federal agencies the responsibilities of leadingcoordination activities for categories of data, for example:

• The Secretary of the Interior heads the FGDC, and the Department ofInterior’s U.S. Geological Survey (USGS) is responsible for basecartographic data and geologic data.

• The Secretary of Commerce heads the Federal Geodetic ControlSubcommittee (FGCS) of the FGDC, and the Department ofCommerce’s National Geodetic Survey (NGS) is responsible forGeodetic Control and Bathymetry, major components of the NationalSpatial Reference System (NSRS).

USGS has responsibilities for standards related to base cartographic andgeologic information. NGS has responsibilities for standards related togeodetic control and bathymetry. Other Federal agencies (U.S. ForestService, National Park Service, Bureau of Land Management, FEMA,Natural Resources Conservation Service, etc.) have similarresponsibilities for cadastral, transportation, soils, vegetation, wetlands,floodplain mapping, etc.

By participating and encouragingothers to participate in the NSDI,the various Federal agencies canrealize several opportunities forcarrying out their missions.Making their data availablethrough the NSDI increases theopportunities for these data to beused in decisions made at the local, regional, national, and global scales,and it helps to increase the relevance of Federal activities. Through theNSDI, responsible Federal agencies can locate data produced by othersthat can supplement their data collection efforts, and they can identify

The intent is to avoidduplication of effort andprovide accurate and up-to-date spatial data most cost-effectively to all.

APPENDIX


organizations that are candidates for collaborative data collection anduse.

A major component of the NSDI is the development and implementationof a national digital geospatial data framework. Although applicationsof digital geospatial data vary greatly, users have recurring needs for afew common themes of data. These data themes (the framework)include orthoimagery, elevation, transportation, hydrography, politicaland administrative boundaries, cadastral, and geodetic control.

The recurring needs for these data themes are not being met consistentlybecause of limited investment, gaps in coordination, and a lack ofcommon approaches. As a result, important information is not availablefor many areas, and multiple organizations support duplicate dataactivities for other areas. Because no coordinated mechanism exists tomaintain and manage the common data being collected by the public andprivate sectors, costs are higher, and efficiency is reduced for individualorganizations, as well as for the Nation.

The purpose of the framework concept is to organize and enhance,throughout all levels of the government and the private sector, thecollection, maintenance, and dissemination of basic, consistent digitalgeospatial data. The framework will facilitate data sharing and provide abase on which an organization can accurately register and compile otherthemes of data or add application-specific information. Sharedcollection and maintenance will reduce expenditures for data collectionand integration, allow organizations to focus on their primary business,expand the user base for data being collected, and increase dataavailability over broader geographic areas.

8.2 National Spatial Reference System (NSRS)

NGS’ document, National Geodetic Survey: Its Mission, Vision, andStrategic Goals, was prepared to describe to the public itsimplementation of the NSRS, mapping and charting activities, research,public outreach, and other related activities.

As currently envisioned, the NSRS framework, when completed in all 50states, would include some 16,000 geodetic control stations. Becausethis number of stations is impractical for NGS to establish and maintainwith present funding and human resources, NGS has separated these

APPENDIX


stations into categories for the purpose of assigning responsibility forestablishment and maintenance.

• The Federal Base Network (FBN) consists of a very high-accuracy,four-dimensional (latitude, longitude, orthometric height, and time)network of over 1,400 monumented stations with 100-km nominalspacing throughout the U.S. and its territories. The FBN will alsocontain additional stations as needed in areas of subsidence or crustalmotion, or as needed in support of Federal aircraft navigationalrequirements.

• The Cooperative Base Network (CBN) consists of approximately14,600 three-dimensional (3-D) monumented stations, with 25- to 30-km nominal spacing, established and maintained through cooperativeagreements with other Federal, state, and local government agencies.

In order to cost-effectively achieve 2-cm elevation accuracy, DGPSbase stations need to be within 10 km of the rover units. For thisreason, FBN/CBN point spacing of 25-30 km is inadequate. Anominal spacing of 10 km is recommended.

• The User Densification Network (UDN) consists of additionalmonumented stations, connected to the FBN or CBN in accordancewith FGCS Standards and Specifications, which contribute to thepublic good. These are normally surveyed by the private sectorwithout a cooperative agreement.

To facilitate the User Densification Network, user-friendly “bluebooking” programs are required for import of processed GPS filesand monument descriptions.

The data management, archiving, and dissemination for both FBN andCBN is the responsibility of NGS. The NGS responsibility for the UDNis to act as the depository and purveyor of control survey data, providedthe surveys were performed by or for National, state, or localgovernments and satisfy rigorous accuracy requirements.

APPENDIX


8.3 Vertical Datums

All measurements made bysurveyors to determine anddepict horizontal positions andelevations should relate to a“reference datum.” The NorthAmerican Datum of 1983 (NAD83) has replaced the NorthAmerican Datum of 1927 (NAD27) as America’s standardhorizontal datum. Similarly,the North American VerticalDatum of 1988 (NAVD 88) has replaced the National Geodetic VerticalDatum of 1929 (NGVD 29) as America’s standard vertical datum.However, full implementation may never occur, and current deficiencieswill worsen, without sustained Federal funding for implementation.

8.3.1 Local Mean Sea Level

America's initial use oflocal mean sea level as avertical datum referencewas based on the readilyobserved tidal cycles ofmean hourly waterelevations observed overa 19-year Tidal Datum.

The arithmetic meanof these observationsprovided the level usedas local mean sea level.However, there are manyvariables that affect thedetermination of local mean sea level, and an attempt was made in the1920s to define a consistent nationwide vertical datum to replace theconfusing morass of local datums at every seaport.

After a decade, the NorthAmerican Vertical Datum of1988 (NAVD 88) has still notbeen implemented in majorportions of the U.S. In fact,dozens of diverse verticaldatums are in use throughoutthe U.S., with unknownlinkage to NAVD 88.

Figure 21 Vertical Datum Reference Options

APPENDIX


8.3.2 National Geodetic Vertical Datum of 1929 (NGVD 29)

The National GeodeticVertical Datum of1929 (NGVD 29)combined a series ofprecise levelingsurveys, referenced to21 tide gages in theU.S. and five inCanada. The object ofNGVD 29 was toprovide a fixed datumthat was supposed tobring a consistentrelationship to all vertical determinations in the U.S. Until recently,NGVD 29 has been the vertical datum for all maps of the U.S. producedby the U.S. Geological Survey (USGS) and other Federal agencies. Asnewer data were incorporated into NGVD 29, surveyors and mappersbecame dissatisfied with the inconsistencies in NGVD 29. Since theadvent of GPS, the requirements for change have become obvious, andthe technical means are available, affordable, and relatively risk-free.

8.3.3 North American Vertical Datum of 1988 (NAVD 88)

The North American Vertical Datum of 1988 (NAVD 88) was necessaryto remove the inconsistencies and distortions in the NGVD 29. Forexample, first-order elevations leveled from zero NGVD in New Yorkwere significantly different from first-order elevations leveled from zeroNGVD in Philadelphia. A decision was made by NGS, and itscounterpart agencies in Canada and Mexico, to adopt a vertical datumbased on a mathematical surface that closely approximates the geoid.Approval and funding to establish the new datum was received in 1978.International in scope and definition, NAVD 88 was constructed from1.3 million kilometers (800,000 miles) of precise leveling measurementsthroughout Mexico, the United States, and Canada, and thereadjustments of about 600,000 permanent benchmarks. In 1998 dollars,this represents an investment of well over one billion dollars.

Although a Federal Register Notice of June 24, 1993, designated NAVD88 as the official replacement of the older NGVD 29 datum, it is still

Figure 22 Height Measurements

APPENDIX


acceptable to use NGVD 29 and diverse vertical datums (e.g., Mean LowLow Water (MLLW) Tidal Datums, the 1974 Low Water ReferencePlane (LWRP) for the Lower Mississippi River, the International GreatLakes Datum of 1985 (IGLD85), local construction project datums, orother reference planes established by local jurisdictions), provided therelationship of such datums to NAVD 88 is clearly noted andunderstood. Unfortunately, the distinctions are quite complex and areoften misunderstood, and many jurisdictions totally disregard thisrequirement.

A major problem resultingfrom conversion from NGVD29 to NAVD 88 is that nearlyall contour lines on USGSand other maps are obsolete.These contour lines werephotogrammetrically compiledto NGVD 29, and, withtraditional photomappingtechnology, it is veryexpensive to recompile thesecontours to NAVD 88. Theleast expensive solution is toplace a correction note oneach map, saying for examplethat "all NAVD 88 elevationsare 2.3 feet lower than theNGVD 29 elevations shownon this map." In other words,the user would need toconvert the 10-foot contourline to 7.7 feet, the 20-footcontour line to 17.7 feet, etc.The adjoining map, however,may have a correction of 2.2 feet instead of 2.3 feet. Therefore, at theedge of adjoining maps, where NGVD 29 10-foot contour linespreviously joined, one map would convert this contour line elevation to7.7 feet, and the adjoining map would convert this elevation to 7.8 feet.This presents major problems for users, especially FEMA.

The problems discussed here didnot have practical solutions until1997 when modern aerial surveyfirms proved that they could cost-effectively acquire high-resolution and high-accuracyDigital Elevation Models (DEMs)nationwide using airborne GPSrelated to NAVD 88, inertialmeasuring units (IMUs), andlaser and radar sensors toaccurately survey the terrain.With such DEMs, GIS specialistswould know both the NGVD 29and NAVD 88 orthometricheights of billions of data pointsnationwide, and could usecomputers to compile accuratecontour lines at any desiredcontour interval and to anyexisting or future vertical datum.

APPENDIX


8.4 Relative vs. Absolute Accuracy

The accuracy of orthometric heights has traditionally depended on:

1. Starting Point: The accuracy of benchmarks (or 3-D surveymonuments) used to extend orthometric heights inland from theoceans

2. Procedures: The survey equipment and procedures used

3. Route: The route taken to survey from the benchmark to the newpoint(s) for which orthometric heights are required.

All three of these traditional dependencies contribute to elevationproblems worldwide.

• Surveyors have long known that NGVD 29 elevations, for benchmarksand 3-D survey monuments, were not sacrosanct. Surveyors founderrors and inconsistencies on a daily basis, but little could be done.Since points are surveyed relative to benchmarks that have elevationerrors (often of unknown magnitude), all orthometric heights aredetermined with relative accuracy, and not absolute accuracy.

• Surveyors have long used "orders" and "classes" to categorize theaccuracy of their surveys. First-order, Class I surveys were universallyknown as the best, and surveyors strove to achieve perfect verticalcontrol surveys with zero misclosures. Here, too, the accuracy oforthometric heights are relative accuracy, because each survey "order"and "class" allows specified errors relative to the distance surveyed.

• Geodesists, but few others, recognized that orthometric heights variedaccording to the route used to survey from a known benchmark to anunknown point. This is because local variations in gravity (caused bymass excesses or deficiencies in the Earth’s crust) cause the surveyor’splumb bob to point slightly away from the center of the Earth invariable directions. Orthometric heights, relative to diverse traverse orleveling routes, are bewildering to the survey profession. Until GPSand modern gravimeters enabled geodesists to model the geoid, therewas no effective way to account for local variations in gravity.

Figure 23 provides an example. For illustration purposes, supposepoints A, B, C, and D are on the north, south, east, and west sidesrespectively of a large, still lake; and suppose points A and B are both atthe water’s edge. Point C is east of the lake, on a mountainside. Point D

APPENDIX


is west of the lake, in a valley. One surveyor surveys from A to C to B,and then back to C and A, with zero misclosure; the elevation of point Bis determined to be higher than A. Another surveyor surveys from A toD to B, and then back to D and A, also with zero misclosure; but theelevation of point B is determined to be lower than A. Is B higher thanA, or lower? Both answers cannot be right -- or can they?

In fact, it is likely that the two elevations at B will differ because of amass excess (mountain) near point C, and a mass deficiency (valley) nearpoint D. The two surveyors did nothing wrong, yet they ended up withtwo different orthometric heights for point B, both relative to point A;and both their surveys had zero misclosures. Because water surfaces areequipotential surfaces (having equal forces of gravity), the elevations(technically, the “dynamic heights”) of points A and B, by definition,must be equal, but this can be verified only if the surveyor surveysdirectly over the level (equipotential) surface of the lake between A andB, rather than surveying around the east or west sides of the lake wherethe directions of gravity are different.

• Precise elevation surveys are routinely performed to monitor landsubsidence, intended to ensure the supply of water for humanconsumption, ranching, and agriculture. In central California, highaccuracy leveling erroneously concluded that prior subsidence hadstabilized and that some ground was actually rising in elevation.However, GPS surveys from distant Continuously Operating ReferenceStations (CORS) proved that the benchmark used for the leveling wassubsiding also, at a faster rate than some other points leveled relativeto this benchmark. This made some points appear to be rising, relativeto the benchmark, when they were in fact continuing to subside.

Lake

Valley

Mountain

A

B

D

C

Figure 23 Example of Route-Dependent Elevations

APPENDIX


• Since the advent of the Global Positioning System (GPS) anddifferential GPS (DGPS), surveyors now determine 3-D coordinateswith absolute accuracy of several centimeters (2.54 cm = 1 inch) inmost locations. Although DGPS is technically a relative surveyprocess, the difference is that GPS surveys can be relative to NGS’CORS that are surveyed so well that their absolute errors areconsidered to be zero, relative to the Earth’s center. Thus, GPSsurveys conducted relative to CORS yield absolute accuracy.

8.5 NAVD 88 Implementation

While NAVD 88 was designed and made official over seven years ago,its implementation into the NSRS has yet to be achieved, and itsdeficiencies in certain regions threaten its very existence. In California,only 30% of the existing vertical control monuments were ever includedin NAVD 88, because of their uncertain stability. Of this network of17,000 NAVD 88 monuments, subsidence or seismic activity has nowsignificantly disturbed 25%. Similar subsidence issues exist in Texas,Louisiana, Florida, South Carolina, North Carolina, Virginia, NewJersey, New York, Michigan, and Minnesota. The National Academy ofSciences has estimated subsidence damage costs in each of these statesto exceed $10 million annually.

Relatively small changes in elevation often have profound impact. Forexample:

• The National Research Council conservatively estimated the annualcosts due to increased flooding and structural damage from subsidencein the U.S. to be in excess of $125 million.

The California Department of Water Resources estimates aquifers in theState are over drafted by 10 million acre-feet annually, an increase of75% over the last five years.

8.6 Global Positioning System (GPS)

America’s Global Positioning System (GPS) already is the world’snext Utility System. As shown by the chart below, the FreedoniaGroup and similar organizations project a meteoric rise in GPS usersover the next decade, with communications, automotive, and otherapplications impacting virtually every American. During the past few

APPENDIX


years, the National Geodetic Survey (NGS) and others have proventhat GPS can provide good 3-D positioning, with accurateheights/elevations, and not just 2-D positioning. Airborne GPS,combined with inertial, laser, and radar sensors, also enable Americato satisfy its decades-long requirements for accurate Digital ElevationModels (DEMs) vital to thousands of Federal, state, county, localcommunities, and private organizations that rely on modernGeographic Information System (GIS) technology for competitivebusiness practices.

Requirements - PhilosophyFuture GPS User Sectors - $M

(Based upon Freedonia Group Report, Cleveland, Ohio - 1997)

0

500

1000

1500

2000

2500

3000

3500

4000

1996 2001 2006

CommunicationAutomotiveGov./Inst.Civil AviationM aritimeIndustrialRecreationM ilitaryAgricultureOtherAerospace

Sales of GPS Equipment & Services

Trying to figure out where he is and where he is going is probably one ofman’s oldest problems. Navigation and positioning are crucial to somany activities, and yet the process has always been quite cumbersome.Over the years, all kinds of technologies have been tried to simplify thetask, but each has had some disadvantage.

• Landmarks work only in local areas. They are subject tomovement or destruction by environmental factors.

• Dead reckoning is very complicated. Accuracy depends onmeasurement tools that are usually relatively crude. Errorsaccumulate quickly.

• Celestial navigation is very complicated. It works only at night,in good weather.

Figure 24 Projected Growth in GPS Users

APPENDIX


• OMEGA has limited precision. It is based on relatively fewradio direction beacons. Accuracy is limited and subject to radiointerference.

• LORAN has limited coverage (mostly coastal). Accuracy isvariable and is affected by geographic situation. LORAN is easyto jam or disturb.

• SatNav is based on low-frequency Doppler measurements, so itis sensitive to small movements at the receiver. There are fewsatellites, so updates are infrequent.

Ultimately, the U.S. Department of Defense developed a technicalsolution to satisfy operational requirements worldwide. The result is theGlobal Positioning System (GPS), a system that has changed navigationand positioning forever. Its military value was proven in 1991 during theMid-East War when GPS-guided cruise missiles and “smart bombs”contributed greatly to the nation’s success. Today, civil GPS users faroutnumber military users, and GPS is equally successful for a myriad ofinnovative applications.

GPS is a worldwide radio-navigation system formed from a constellationof 24 satellites and their ground stations. GPS uses these "man-madestars" as reference points to calculate positions accurate to a matter ofmeters. In fact, with advanced forms of GPS, it is possible to makemeasurements to better than a centimeter! In a sense, it is like givingevery square foot on the planet a unique address.

GPS receivers have been miniaturized to just a few integrated circuitsand so are becoming very economical, which makes the technologyaccessible to virtually everyone. These days, GPS is finding its way intocars, boats, planes, construction equipment, movie making gear, farmmachinery, even laptop computers. Soon, GPS will become almost asbasic as the telephone. Indeed, GPS is quickly becoming a universalutility.

• Name: NAVSTAR Manufacturer: Rockwell International

• Altitude: 20,000 Kilometers

• Weight:1900 lbs. (in orbit)

• Size:17 ft with solar panels extended

• Orbital Period: 12 hours

APPENDIX


• Orbital Plane: 55 degrees to equatorial plane

• Planned Life Span: 7.5years

• Current constellation: 24Block II productionsatellites

• Future satellites: 21 BlockIIR’s developed by MartinMarietta.

Ground Stations (also knownas the "Control Segment")monitor the GPS satellites,checking both theiroperational health and theirexact position in space. Themaster ground stationtransmits corrections for thesatellite's ephemerisconstants and clock offsetsback to the satellitesthemselves. The satellitescan then incorporate theseupdates in the signals theysend to GPS receivers.

There are five monitorstations: Hawaii, AscensionIsland, Diego Garcia,Kwajalein, and ColoradoSprings.

The GPS provides two levelsof service -- a Standard Positioning Service (SPS) for general public useand an encoded Precise Positioning Service (PPS) primarily intended foruse by DOD. The SPS provides civil users a worldwide service withoutcharge or restrictions. The SPS accuracy is intentionally degraded by theDOD through the use of a time-varying bias called Selective Availability(S/A). SPS (with S/A on) has a predictable accuracy of 100 meters(horizontal) and 156 meters (vertical). If S/A is turned off by DOD, the

How GPS works in five steps:

�� The basis of GPS is"triangulation" from satellites.

�� To "triangulate," a GPS receivermeasures distance using thetravel time of radio signals.

�� To measure travel time, GPSneeds very accurate timing thatit achieves with atomic clocks.

�� Along with distance, you need toknow exactly where the satellitesare in space. High orbits andcareful monitoring are thesecret.

�� Finally you must correct for anydelays the signal experiences asit travels through theatmosphere.

Improbable as it may seem, thewhole idea behind GPS is to usesatellites in space as reference pointsfor locations here on earth. By veryaccurately measuring our distancefrom four satellites, we can"triangulate" our positionanywhere on earth.

APPENDIX


predictable positioning accuracy of a single GPS receiver will improveto 22 meters (horizontal) and 28 meters (vertical) -- still inadequate formany civil requirements.

8.7 Differential GPS (DGPS)

Basic GPS is the most accurate radio-based navigation system everdeveloped, and for many applications it is plenty accurate. DifferentialGPS or "DGPS" can yield meter-level accuracy in moving applicationsand centimeter-level accuracy in stationary situations. That improvedaccuracy has a profound effect on the importance of GPS as a resource.With it, GPS becomes more than just a system for navigating boats andplanes around the world. It becomes a universal measurement system,capable of positioning things on a very precise scale.

Differential GPS (DGPS) involves the cooperation of two receivers, onethat is stationary (at a known point) and another that is roving aroundmaking position measurements (at unknown points). The stationaryreceiver is the key. It ties all the satellite measurements into a solid localreference. GPS receivers use timing signals from at least four satellitesto establish a position. Each of those timing signals is going to havesome error or delay depending on what sort of perils have befallen it onits trip down to earth. Since each of the timing signals that go into aposition calculation has some error, that calculation is going to be acompounding of those errors. The satellites are so far out in space thatthe little distances traveled here on earth are insignificant. If tworeceivers are fairly close to each other, within a few hundred kilometers,the signals that reach both of them will have traveled through virtuallythe same slice of atmosphere, and so will have virtually the same errors.Thus, errors detected at the DGPS base station (known point) are used tocorrect the positions of the unknown points, simultaneously observingthe same four (or more) GPS satellites.

Soon DGPS may be able to resolve positions that are no farther apartthan the width of one’s little finger. Automatic construction equipmentwill be able to translate CAD drawings into finished roads without anymanual measurements. Self-guided cars will take “drivers” across townwhile they quietly read in the back seat. To understand how this kind ofGPS is being developed, it is necessary to understand a little about GPSsignals, and a little about how surveyors are using GPS right now.Surveyors have been using GPS to do extremely precise surveys for

APPENDIX


years, fixing points with relative accuracy on the order of millimeters.Furthermore, GPS technology continues to evolve. GPS satellites of thefuture, with multiple “carrier phase” frequencies are expected to beorders of magnitude more accurate than "code-phase GPS” used today.

8.7.1 DGPS for Air Navigation and Safety

Realizing the capabilities and value of GPS, the Federal AviationAdministration (FAA) is currently implementing a Wide AreaAugmentation System (WAAS) based on GPS and geostationarycommunications satellites. This system will bring the accuracy andreliability of differential GPS across the entire continent and make agreat contribution to aviation safety and air traffic control. The ideagrew out of some very specific requirements that basic GPS could nothandle by itself. It began with "system integrity." GPS is very reliable,but every once in a while a GPS satellite malfunctions and givesinaccurate data. The GPS monitoring stations detect this sort of thingand transmit a system status message that tells receivers to disregard thebroken satellite until further notice. Unfortunately, this process can takemany minutes, whichcould be too late foran airplane in themiddle of a landing.To counter thisanomaly, the FAAwill establish its ownmonitoring systemthat will respond

much quicker. In fact,they figured they couldpark a geosynchronoussatellite somewhere over the earth that would instantly alert aircraftwhen there was a problem. Then they reasoned that they could transmitthis information on a GPS channel so aircraft could receive it on theirGPS receivers and without any additional radios. Adding anothersatellite helps with positioning accuracy, and it ensures that plenty ofsatellites are always visible around the country!

The FAA figures that with about 24 reference receivers scattered acrossthe United States, they can gather good correction data for most of thecountry. This data will make GPS accurate enough for "Category 1"

GroundStation

Monitor StationNetwork

GPS

Differential Corrections

Figure 25 Operation of the FAA Wide AreaAugmentation System (WAAS)

APPENDIX


landings (i. e., very close to the runway but not zero visibility). Thissystem is currently being implemented.

To complete the system, the FAA will establish "Local AreaAugmentation Systems" (LAAS) near runways. These will work like theWAAS, but on a smaller scale, and perhaps with pseudolites instead ofgeosynchronous satellites. The reference receivers will be near therunways, and so will be able to give much more accurate correction datato the incoming planes. With a LAAS, aircraft will be able to use GPSto make Category 3 landings (zero visibility). For this to work, heightdata are critical to within a few centimeters. America’s height referencesystem, NAVD 88, is crucial to the success of LAAS and the ultimatesafety of millions of air travelers in the future.

However, other users of DGPS augmentation systems cannot rely onthese FAA systems to satisfy ground positioning requirements for thefollowing reasons:

• “This signal reception from the WAAS geostationary satelliterepresents a serious concern for users on the surface where there is apotential for multipath, shadowing, and blockage of WAAS signalsdue to natural and manmade obstructions.26” Essentially, ageosynchronous satellite designed for optimum communications within-flight aircraft would not satisfy ground users who regularly havepoor visibility to a geosynchronous satellite “parked” over the equator.

• Similarly, LAAS communications would be designed to optimizeaviation safety, using aviation frequencies. LAAS DGPS correctionswould certainly not focus on the need to penetrate buildings andmountains, as are NDGPS signals. Furthermore, it may not be inanyone’s best interest to make compromises that could jeopardizesafety for either air or ground travelers.

8.7.2 DGPS for Land Navigation and Vehicle Tracking

Navigation is the process of getting something from one location toanother; tracking is the process of monitoring it as it moves along.Commerce relies on fleets of vehicles to deliver goods and services

26 Arnold, James A., Federal Highway Administration, March 24, 1998, “GeostationarySatellite Coverage”, Nationwide DGPS Report, pp. 95-105.

APPENDIX


either across a crowded city or through nationwide corridors. Soeffective fleet management has direct bottom-line implications, such astelling a customer when a package will arrive, spacing buses for the bestscheduled service, directing the nearest ambulance to an accident, orhelping tankers avoid hazards.

GPS used in conjunction with communication links and computers canprovide the backbone for systems tailored to applications in the areas ofagriculture, mass transit, urban delivery, public safety, and vessel andvehicle tracking. Consequently, police, ambulance, and fire departmentsare adopting systemslike Trimble’s GPS-based AVL (AutomaticVehicle Location)Manager to pinpointboth the location of theemergency and thelocation of the nearestresponse vehicle on acomputer map. Withthis kind of clear visualpicture of the situation,dispatchers can reactimmediately and confidently.

The goal of public safety agencies is to make every emergency responsetime as fast as possible. When an emergency erupts, an extra second ortwo can mean the difference between a life saved and a life lost. Adispatcher must make vital decisions in even the best circumstances.The task becomes even more complex in a metropolis like Chicago, witha tangled web of streets and fifteen million emergency calls each year.To help make a difference, Chicago has developed an emergencyresponse system built on GPS based Automatic Vehicle Location (AVL)hardware and software.

Under the old system, Chicago’s dispatchers had to make decisions basedon data typed on 3”x5” cards. Under the new system, they have all thevital information displayed on the digital maps right in front of them.Dispatchers see the entire city on a digital map allowing 911 calls to bepinpointed instantly. Emergency response vehicles are displayed asicons so the most available unit can be determined. Dispatchers can evenzoom in to see building footprints, addresses, even building height, type,

DGPS

StandardPositions

AddCorrections

DifferentiallyCorrectedPositions

Figure 26 GPS Configuration for AutomatedVehicle Tracking

APPENDIX


and access. This new GPS-based dispatch system displays a wealth ofother information, including location of fire hydrants, street directions,and street width. By taking advantage of GPS and by having the tools toimmediately identify the best unit to respond, dispatchers will be able totrim seconds (if not minutes) off of response times. A year from now,there will be Chicago citizens alive who simply would not be alivewithout this new system.

8.8 Nationwide DGPS (NDGPS)

In the early days of GPS, private companies with big projects demandinghigh accuracy -- groups like surveyors or oil drilling operations --established their own DGPS reference stations. This is still a verycommon approach, using a reference receiver and setting up acommunication link with your roving receivers. The U.S. Coast Guard(USCG), U.S. Army Corps of Engineers (USACE), and someinternational agencies are establishing reference stations all over theworld, especially around popular harbors and waterways. These stationsoften transmit on the radio beacons that are already in place for radiodirection finding (usually in the 300 kHz range). Anyone in the area canreceive these corrections and radically improve the accuracy of theirGPS measurements. Most ships already have radios capable of tuningthe direction finding beacons, so adding DGPS will be quite easy. Manynew GPS receivers are being designed to accept corrections, and someare even equipped with built-in radio receivers.

For maritime safety along U.S. coastlines, the Great Lakes, and theMississippi River, the USCG and USACE jointly operate DGPS basestations with low frequency/medium frequency (LF/MF) radio beacons thattransmit differential GPS corrections in real time to users. At the beginningof 1998, the USCG’s DGPS Navigation Service had 54 such DGPS stationsin operation (44 in the continental U.S., seven in Alaska, two in Hawaii, andone in Puerto Rico). These DGPS beacon sites, which broadcast differentialcorrections up to 250 miles away, are extremely effective for multiple users,not just maritime users. In fact, before implementing their system 24 hoursper day, the USACE would receive complaints from Midwest farmerswithin hours of turning off a DGPS station.

APPENDIX


Figure 27 Current USCG/USACE DGPS Network

With requirements from the Federal Railroad Administration (FRA) for

Figure 28 Proposed Nationwide Differential GPS (NDGPS)

APPENDIX


Positive Train Control, and demands from the farm states for PrecisionFarming, the U.S. Congress provided $2.4 million in FY98 funds tobegin the USCG’s expansion into a Nationwide DGPS. The $2.4 millionprovided funding for reusing and retrofitting six of 66 Ground WaveEmergency Network (GWEN) systems, being decommissioned by the U.S.Air Force, for conversion into DGPS beacon sites. The first suchGWEN/NDGPS conversion, in Appleton, Washington, was a total success.

When fully funded, some of the 66 GWEN sites will be converted in placeinto NDGPS sites, while others will be relocated to other States. Of thistotal, 54 new NDGPS sites will be in the continental U.S., with 12 more inAlaska—for a total of 120 NDGPS sites. This is one of the largest defense-to-civil conversions in history, and DGPS users throughout America canthank forward-looking innovators in the Federal DOT, U.S. Coast Guard,and Federal Railroad Administration for bringing this capability to the entirecountry.

When 120 total NDGPS sites are operational, the United States will haveredundant DGPS coverage nationwide provided free to all constituents, withavailability of 99.999%, continuous integrity monitoring by the USCG toensure that the system is operating effectively, and nonproprietary standardscurrently used by dozens of other countries.

8.9 Digital Elevation Models (DEM)

The most persistent requirement expressed by users, from all application,was for high accuracy DEMs, based on a standard vertical datum --NAVD 88. This section explains the various DEM types available, andtheir comparative accuracies.

A DEM is the digital cartographic representation of the elevation of theland (z value) at regularly spaced intervals in x and y directions (eastingsand northings, or longitude and latitude). This definition of DEM alsoapplies to Digital Terrain Elevation Data (DTED) produced by theNational Imagery and Mapping Agency (NIMA). The term “DEM” issometimes used generically to mean the digital cartographicrepresentation of the earth in any form, rectangular grids or lattices,triangular networks, or irregular spot heights and break lines, forexample.

APPENDIX


8.9.1 USGS DEM Types

USGS produces several standard types of DEM Data:

• 7.5-minute DEMs normally have 30- by 30-meter point spacing, usingUniversal Transverse Mercator (UTM) coordinates on the NAD 27 orNAD 83 horizontal datum and NGVD 29 vertical datum. Theyprovide the same coverage as standard USGS 7.5-minute quadrangles(1:24,000-scale map), covering 7.5 minutes of latitude by 7.5 minutesof longitude. Coverage of the entire United States is complete. USGSnow offers a 7.5-minute DEM having 10- by 10-meter point spacing.This product is produced when USGS has a cooperator (funding,workshare, Innovative Partnership, etc.)

• 30-minute DEMs have 2- by 2-arc second point spacing –approximately 60- by 60-meter point spacing in x and y – withgeographic coordinates (latitude/longitude) on the NAD 27 or NAD 83horizontal datums and NGVD 29 vertical datum. With half thecoverage of a standard USGS 30-minute by 60-minute quadrangle(1:100,000-scale map), it takes two 30-minute DEMs to cover the areaof a 1:100,000-scale quad. Coverage of the United States isincomplete.

• 1-degree DEMs have 3- by 3-arc second point spacing –approximately 100- by 100-meter spacing in x and y – with geographiccoordinates on the World Geodetic System (WGS) 72 or WGS 84horizontal datum and NGVD 29 vertical datum. They providecoverage in 1- by 1-degree blocks. They are formatted to USGSspecifications from NIMA’s DTED. Coverage of the entire UnitedStates is complete.

• Alaska has several other types of DEMs, with point spacing thatadjusts for the convergence of meridians at northern latitudes.

8.9.2 USGS DEM Levels and Their Vertical Accuracy

DEMs are classified into one of three levels of quality. There are varyingmethods of data collection and degrees of editing available for DEMdata. All USGS DEMs are tested and assigned a vertical root meansquare error (RMSE).

• Level 1 DEM data are created by automated stereo correlation ormanual profiling from aerial photographs such as photos from the

APPENDIX


National High Altitude Photography (NHAP) program, initiated in1980, which produced images at a scale of 1:80,000, and the follow-onNational Aerial Photography Program (NAPP), initiated in 1987,which produces images at a scale of 1:40,000. Level 1 30-minuteDEMs may be derived or resampled from Level 1 7.5-minute DEMs.The Level 1 7.5-minute DEMs meet the minimum accuracyrequirements to support production of Digital Orthophoto Quarter-quads (DOQs) compiled at a scale of 1:12,000.

• Level 2 DEMs are created from 1:24,0000-scale contours for theconterminous U.S. and from 1:63,000-scale contours in Alaska. Level2 DEMs are elevation data sets that have been processed or smoothedfor consistency and edited to remove identifiable systematic errors.

DEM data derived from hypsographic and hydrographic datadigitizing, either photogrammetrically or from existing maps, areentered into the Level 2 category after review on a DEM editingsystem. An RMSE of one-half contour interval is the maximumpermitted with no errors greater than one contour interval. Theaccuracy and data spacing are intended to support computerapplications that analyze hypsographic features to a level of detailsimilar to manual interpolations of information from printed sourcemaps.

DEM Level 1 DEM Level 2

Figure 29 Comparison of Level 1 and Level 2 DEMs

APPENDIX


• Level 3 DEM data are created from DLGs that have been verticallyintegrated with all categories of hypsography, hydrography, ridge line,break line, drain files, and all survey control networks. They require asystem of logic incorporated into the software interpolation algorithmsthat clearly differentiates and correctly interpolates between thevarious types of terrain, data densities, and data distribution. AnRMSE of one-third of the contour interval is the maximum permitted,with no errors greater than two-thirds contour interval.

Level 2 and Level 3 DEM data derived from contours generally representslope more accurately than Level 1 DEM data. The USGS does notcurrently produce Level-3 DEM data.

8.10 Photogrammetrically Generated DEMs

GPS technology has dramatically impacted the photogrammetryprofession, because airborne GPS and Inertial Measuring Units (IMUs)are now used for navigation in the aerial photography process, and theygreatly simplify the aerial triangulation process required forphotogrammetric mapping, contouring, and/or DEM generation.However, to achieve 6" accuracy for DEMs, the photogrammetricmapping aircraft needs to fly at a very low altitude to acquire large-scalephotography, and production costs increase significantly with increasedrequirements for vertical accuracy. DEMs available today from USGSdo not satisfy many modern requirements for DEMs, as summarized inTables 1 through 5 in Chapter 3 of this report, but they do represent whatis practical and affordable using state-of-the-art photogrammetrictechniques.

Detailed procedures for DEM extraction, editing, matching, and qualitycontrol are detailed in Chapter 6 of Digital Photogrammetry, AnAddendum to the Manual of Photogrammetry, published in 1996 by theAmerican Society for Photogrammetry and Remote Sensing (ASPRS).Page 140 explains fundamental problems with photogrammetriccompilation of DEMs in vegetated areas, and the potential of airbornelaser scanners (see LIDAR below) to solve such problems.

APPENDIX


8.11 LIDAR-Generated DEMs

A recent article by DeLoach27 discusses photogrammetry as well as LIDAR andIFSAR, which have recently become viable alternatives for generating DEMs.Airborne LIDAR uses an aircraft equipped with airborne GPS and InerrtialMotion Unit (IMU) sensors for 3-D positioning and orientation in airspace, plusa laser system designed to measure the 3-D coordinates of passive targets. Thelaser measures the range to the ground surface (or targets), and, when combinedwith the position and orientation of the aircraft, yields the 3-D positions of thetargets, accurate to approximately 15 cm (6 inches) at flying heights of 6,000feet. LIDAR data are typically three or more times more dense thanphotogrammetrically captured elevation data, and they provide an ideal DEMfor the rectification of digital orthophoto images. LIDAR yields verticalaccuracy of 1-2 feet from flying heights of 20,000 feet. The better LIDARsystems can capture both the first and last returns from each laser pulse,providing both tree canopy elevations and ground elevations.

27 DeLoach, Stephen, “Photogrammetry: A Revolution in Technology”, ProfessionalSurveyor, March 1998, pp. 8-14.

Figure 30 Comparison of a Level 2 and LIDAR DEMs

DEM Level 2 LIDAR DEMThe Level 2 DEM was provided by the New York State Department of

Environmental Conservation (NYSDEC). The LIDAR DEM, provided by PARGovernment Systems under contact to NYSDEC, was produced by EagleScan

Inc., using the Digital Airborne Topographic Imaging System (DATIS).

APPENDIX


When produced for large areas (e.g., an entire state), the LIDAR DEMscost approximately $500 per square mile for unrestricted use. They arefar superior to either Level 1 or Level 2 DEMs and are also more costly.The cost of Level 2 DEMs variesbetween $10 and $100 per squaremile, depending upon how costs arecomputed, e.g., whether or not thecost of photogrammetric contouringand hypsographic and hydrographicDLG production is included.

Whereas the prior DEMcomparisons (Figure 29) show the superiority of Level 2 DEMscompared with Level 1 DEMs, Figure 30 shows the superiority ofLIDAR-generated DEMs compared with Level 2 DEMs when zoomedin. IFSAR DEMs, explained in the next section, lie somewhere betweenthe two extremes shown in Figure 30.

8.12 IFSAR-Generated DEMs

Airborne IFSAR is similar to airborne LIDAR, except that IFSAR usesradar interferometry instead of lasers to measure the 3-D coordinate ofpassive targets and can do so in all weather conditions, including totalcloud cover. IFSAR accuracy is approximately 1-2 meters when flownat an elevation of 20,000 feet, approximately 3-meters when flown at40,000 feet. A limited number of LIDAR cross strips can be used toimprove the accuracy of the IFSAR DEMs to a few feet.

IFSAR DEMs cost between 1/10th

and 1/20th as much as LIDARDEMs, i.e., between $25 and $50per square mile when mass producedover large areas. With minimaldevelopmental activity, it would bepossible to fly IFSAR and LIDARsensors concurrently in the samemapping aircraft to combine theadvantages of each sensor.

IFSAR DEMs (flown withnorth-south flight lines, forexample) can be significantlyimproved with a few LIDARcross-strips (flown with east-west flight lines) to correctsystematic errors in theIFSAR data.

The estimated cost ofLIDAR DEMs nationwidewould be approximately $2billion, but the estimatedbenefits would be evengreater, as indicated inTable 22.

national height modernization studynational height modernization study national geodetic surveyi...

Documents